Re: Scheduled for reindexing STAS

[Andreas J. Koenig]

> On Sun, 24 Jun 2012 20:40:27 +0200, Steffen Mueller  
> said:

 sm> Hi Andreas,

 sm> I think I need your help on this one.

 sm> On 06/24/2012 08:18 PM, Perl Authors Upload Server wrote:
 >> According to a request entered by Steffen Müller the
 >> following files have been scheduled for reindexing.
 >> 
 >> $CPAN/authors/id/S/ST/STAS/Env-C-0.08.tar.gz
 >> 
 >> Estimated time of job completion: Sun Jun 24 19:19:00 2012 GMT

 sm> This didn't work. STAS was only co-maint previously. I tried to lose
 sm> the co-maint bit and then reindex to make him primary maintainer
 sm> again. Did not work.

 sm> Here's Stas' original email:
 >> Michael Schout / MSCHOUT wants to take over the ownership of Env::C.
 >> I used to be the owner and then passed it to someone else, but
 >> somehow that someone else dropped the ownership w/o re-assigning it
 >> to anybody else, so the module as an orphan now, and I'm listed as
 >> the co-maint, but I can't change permissions.
 >> 
 >> So can you please set Michael as the owner?

I think it worked. When I looked at pause, Stas had "first-come"
ownership and I could pass it to MSCHOUT in the browser in the change
permissions section as normal.

Michael, you should be able to upload now and get indexed. Let us know
if anything does not behave as expected.

-- 
andreas

Re: PAUSE LWP broken on HTTPS?

[Andreas J. Koenig]

> On Tue, 10 Jan 2012 10:48:32 -0500, David Golden  said:

  > Thanks.  I've forwarded this to Andreas in case he isn't monitoring
  > the modules list closely.

Thanks David...

  > On Tue, Jan 10, 2012 at 10:35 AM, Mark Allen  wrote:
 >> Hey I was tailing the log file this morning while I was uploading a new
 >> version of a module, and I noticed that the LWP on the PAUSE server seems to
 >> fail opening on https:// style URLs (say to github) - I suspect that maybe
 >> LWP::Protocol::https isn't installed on the server?

... and thank you, Mark!

Crypt::SSLeay was missing and I installed it now. It seems this was the
only piece missing. Let mem know if it still doesn't download from
github as you'd expect.

Sorry for the inconvenience,
-- 
andreas

Re: PAUSE "Safe" limitations on parsing module version

[Andreas J. Koenig]

> On Sat, 10 Dec 2011 09:51:03 +0200, Alex Efros  
> said:

  > Hi!
  > I'm usually declare module version using this way:

  > use version; our $VERSION = qv('0.1.4'); # update POD & Changes & README

  > and PAUSE was able to parse it successfully several months ago,
  > but now it fail on this. I've also tried slightly different way
  > recommended on `perldoc version`:

  > use version 0.77; our $VERSION = version->declare('0.1.5'); # update 
POD & Changes & README

  > and PAUSE failed to parse this one too.

  > How I should declare module version in PAUSE-compatible way in format
  > '0.1.5' (i.e. not 'v0.1.5' and not '0.001005')?
  > The 'v0.1.5' is incompatible with some tools (like g-cpan in Gentoo
  > Linux), and '0.001005' is just ugly and hard to read.

I just recently uploaded CPAN::Testers::ParseReport which has this line:

use version; our $VERSION = qv('0.2.0');

and pause indexed it.

I'm sorry that your variations thereof do not parse as you expect. I'm
surprised myself. It's not easy to navigate through all the variations
of version trickery. The above incantation is what Damian Conway has
reommended in PBP and that is what pause tries to support as long as
possible.

-- 
andreas

Re: What if there was something less than a co-maintainer?

[Andreas J. Koenig]

> On Tue, 15 Nov 2011 13:25:45 -0600, brian d foy  
> said:

  > This is just an idle thought, so I'm not really proposing anything
  > concrete.

  > I tend to do a lot of maintenance where I get other people involved,
  > but I don't really want to give them upload rights for various reasons.
  > Or, more specifically, I would want to limit them to uploading only dev
  > releases.

Everybody can upload dev releases without asking. They do not get
indexed, no matter if they are maintainer or doing a fork.

  > I'd also like them to be able to affect the RT queue.

Frankly, I do not really understand how RT works. I'm in many RT queues
that I cannot imagine how I came in there. There is some hidden gate I
don't know of.

  > Ultimately, I want to reserve a stable distribution release to primary
  > and co-maintainers as we have it now.

  > It's just an idea, and it's not a pressing problem that I'm trying to
  > solve. I was thinking of it more like github's fork and pull request
  > stuff where there's much less risk in participation with people I don't
  > know.

I'd love to see improvement on that front. Neil Bowers has raised my
attention on the matter. I'm not sure if PAUSE needs to be involved for
this or if maybe a well organized metacpan or something could play a
strong role without having access to pause?

-- 
andreas

Re: Mailinglist accounts

[Andreas J. Koenig]

>>>>> On Sun, 13 Nov 2011 08:39:05 -0500, David Golden  said:

  > On Sun, Nov 13, 2011 at 8:20 AM, Andreas J. Koenig
  >  wrote:
 >>  >> Nobody is in the P5P list.
 >> 
 >>  > Since RJBS is the new Pumpking, he should certainly be on the P5P list.

  > I'm sorry.  I misunderstood.  I thought the P5P list was how we
  > managed permissions for Perl releases so that core modules had maint
  > assigned to "P5P".

No, that's the pumpking table and there RJBS has been put a long time
ago when he released some 5.11.something. perl-* distributions are
treated specially in the indexer, kind of "perl is always right".

  > Then adding an authorized releaser meant putting
  > someone on a P5P list rather than adding a pumpking as maint for every
  > module on CPAN.

Being in the pumpking list

https://pause.perl.org/pause/authenquery?ACTION=who_pumpkin

means that you are allowed to upload a distribution that has a name like
perl distributions have. Not being in this list means that indexing on a
file like perl-5.12.5.tar.gz will fail.

Once a pumpking has uploaded a perl distro the indexer treats the
contained modules as belonging to P5P. That's because P5P has a
hardcoded role in the indexer.

  > Put differently, I thought the mailing lists were how we implemented
  > "group" ownership.  I guess not.  Could your remind me what they *are*
  > used for?

The idea to treat perl releases specially predates the idea of group
permissions. I think it was there from the beginning. But > Since 'P5P'
is a userid that has the 'isa_list' property set it should have all the
properties of group permissions. But this has never been tested because
nobody has ever been put into the 'list2user' table for 'P5P' until
today.

So maybe it actually makes sense to put Ricardo in the list. I suppose
that he now can log into pause and choose to become user P5P for giving
away permission bits or doing some other stuff.

-- 
andreas

Re: [Neil Bowers] Fwd: HTTP::Headers::UserAgent

[Andreas J. Koenig]

>>>>> On Fri, 11 Nov 2011 16:01:38 -0600, brian d foy  
>>>>> said:

  > In article <87lirmjsvd@franz.ak.mind.de>, (Andreas J. Koenig)
  >  wrote:

 >> Forwarding a mail from Neil Bowers who says that his mail to modules@
 >> did not arrive there.

  > I saw both of Neil's original messages *shrug*

I also saw them but only shortly after I sent my forwards to the list *shrug*

-- 
andreas

Re: Mailinglist accounts

[Andreas J. Koenig]

>>>>> On Thu, 10 Nov 2011 22:58:56 -0500, David Golden  said:

  > On Thu, Nov 10, 2011 at 10:53 PM, Andreas J. Koenig
  >  wrote:
 >>>>>>> On Tue, 8 Nov 2011 02:34:07 +, Matt S Trout  
 >>>>>>> said:
 >> 
 >>  > How do PAUSE ids get added and removed from these? RJBS isn't currently 
 >> on
 >>  > the P5P list and it rather seems like he should be - and I can't figure 
 >> out
 >>  > how to make that go.
 >> 
 >> Nobody is in the P5P list. List to user mapping is implemented in the
 >> list2user table which has no web interface. I would have to run an
 >> insert statement for that. Now the question is, what do you mean with
 >> "it rather seems like he should be"?

  > I think that's mst being polite or British or something.

  > Since RJBS is the new Pumpking, he should certainly be on the P5P list.

Even if no other pumpking or not pumpking before ever was? I'm such an
ignorant European barbarian from the middle of that "continent", I'll
never learn British politeness.

OK, I give in. RJBS is now the first ever member of the list2user table:

mysql> insert into list2user values ('P5P','RJBS');
Query OK, 1 row affected (0.00 sec)

Vive le roi de pompe:)
-- 
andreas

Re: [Neil Bowers] Fwd: Lingua::EN::Numbers - request for co-maintainer status

[Andreas J. Koenig]

> On Sat, 12 Nov 2011 09:38:06 -0700, "Sean M. Burke"  
> said:

 > To give you a sense of things: on many different days (I stopped
 > counting at three), me "putting out fires" has involved ACTUAL FIRE.

Sean, you cannot imagine how glad I am to hear you talking! I'm sorry
for the horrors you have to go through and I wish you all the best that
they go by soon. No, not soon, NOW. Immediately!

We (modules@) will deal with the permissions for other modules when
somebody is interested. I hope to see you some day back in the
community, re-powered.

All the best again,
-- 
andreas

[Neil Bowers] Fwd: Lingua::EN::Numbers - request for co-maintainer status

[Andreas J. Koenig]

Forwarding a mail from Neil Bowers that he tried to send to modules@ but
which disappeared on the way to the list.

Neil, I've made you co-maintainer of Lingua::EN::Numbers. I know you as
a conscientiously and thoroughly working man and I'm glad you're
persuing the fix of the bug you discovered. I would believe that it's in
the best interest of Sean that you take over the mainteinance of this
module. Thank you very much for doing that.

I've added co-maintainer rights for Lingua::EN::Numbers::Ordinate and
Lingua::EN::Numbers::Years too. If it turns out you don't need them, you
can remove the permission bits for them on the PAUSE interface.

Sean, if you read this and do not agree, this is completely reversible,
please let us know if you would prefer a different solution.

Jonathan Bailey, I'm CCing you too because you are a co-maintainer and
may have the urge to say something on the matter.

Thanks, everybody!



 Start of forwarded message 
From: Neil Bowers 
Subject: Fwd: Lingua::EN::Numbers - request for co-maintainer status
Date: Fri, 11 Nov 2011 20:19:01 +

This one is perhaps slightly less clear cut.

Allison Randal (one of the co-maintainers who took over Pod::Simple from Sean)
commented in email to me:
> He likes to see evidence of sustained contribution, so he
> won't hand it off on the first patch, but with several good patches, and
> demonstrating interest in keeping on top of bugs and new releases,
> there's a good chance he will. I don't know, he might have already
> handed this one off, I only tracked the Pod:: modules. If he did already
> hand it off, the new maintainer hasn't been active, so it still might be
> a candidate for adoption.

So if he's expressed a clear preference somewhere / to someone, then I don't
want to go against that, but on the other hand, I've made quite a bit of effort
to get in touch with him, and no-one else seems to be stepping forward.


Begin forwarded message:

> From: Neil Bowers 
> Subject: Lingua::EN::Numbers - request for co-maintainer status
> Date: 11 November 2011 10:20:46 GMT
> To: modules@perl.org
> 
> Hi,
> 
> I'd like to be granted co-maintainer status on Lingua::EN::Numbers.
> 
> The current version is 1.01, release by Sean Burke 2005-01-06. He took over 
> the module
> from Stephen Pandich and completely rewrote it.
> 
> I wrote a review of modules for spelling out numbers in English:
>   
> http://blogs.perl.org/users/neilb/2011/09/cpan-modules-for-spelling-out-numbers-in-english.html
> 
> Lingua::EN::Numbers came out top, but I found a bug which I fixed. I'd like 
> to release that fix,
> and resolve at least some if not all of the open RT issues (many of which 
> have been fixed anyway).
> 
> I'd also like to address some of the issues/ideas which came up during the 
> review.
> 
> I first emailed Sean two months ago (12th September). I've tried contacting 
> him a number of ways,
> but have had no response. I recently mailed module-authors, and was told he's 
> retired, with a pointer
> to Pod::Simple, which had been handed over by Sean.
> 
> I had email with the three maintainers of Pod::Simple, who said:
>   - they didn't have another way of contacting Sean other than his cpan 
> address
>   - they thought he'd handed over his modules, but perhaps not all of them
>   - Sean may be waiting for more evidence that I'm serious about 
> maintaining the module
> 
> Was co-maintainer status granted to someone else for Lingua::EN::Numbers, and 
> if so,
> can you tell me who please, so I can contact them.
> 
> Neil
> 

 End of forwarded message 

-- 
andreas

[Neil Bowers] Fwd: HTTP::Headers::UserAgent

[Andreas J. Koenig]

Forwarding a mail from Neil Bowers who says that his mail to modules@
did not arrive there.

I've acted on the request and made Neil the primary maintainer of
HTTP::Headers::UserAgent already, so nothing left to do.


 Start of forwarded message 
From: Neil Bowers 
Subject: Fwd: HTTP::Headers::UserAgent  
Date: Fri, 11 Nov 2011 20:08:56 +

Here's the first message I sent.

Begin forwarded message:

> From: Neil Bowers 
> Subject: HTTP::Headers::UserAgent 
> Date: 11 November 2011 09:05:28 GMT
> To: modules@perl.org
> 
> Hi,
> 
> I'd like to be granted maintenance of the module  HTTP::Headers::UserAgent 
> please.
> 
> There are currently two registered authors:
> 
>   John Neystadt (JNEYSTADT) who released version 1.00 in 1998.
> 
>   Ivan Kohler (IVAN) who released a couple of versions,
>   most recently 3.01 in 2002
> 
> In writing a review of such modules, I discovered that there's a CPAN 
> confusion with this module,
> which results in the 1.00 version being returned as the most recent sometimes 
> and sometimes 3.01.
> 
> I had email with ANDK, who said that it looks like it wasn't properly 
> recorded if/when the module
> was handed over. I suggested that the quickest way (ha, more fool me) to 
> resolve this would be
> if I get permission to take over the module, and upload 3.02.
> 
> The 3.01 release is now just a front-end to one of the other User-Agent 
> modules, and the documentation
> makes it clear that you should probably use a different module. I plan to 
> update the documentation
> to point to my review, and give a list of the best modern modules. I've also 
> added a number of tests,
> just in case there are people out there using the module.
> 
> I've now had email with both authors, who've said they're happy for me to 
> take over the module.
> It took quite a while to track down / get a response from John.
> 
> John wrote:
>> I have no object for you to take over the module.
> 
> Ivan said:
>> I wouldn't say it is worth  covering in a contemporary review of similar 
>> modules
>> other than to point out you shouldn't use it.  :)
> 
> Which is exactly what I do.
> I explained what I want to do, to which he replied:
>> Sure, happy to do so if it makes things easier.
> 
> If you do add me, can I suggest that you remove at least John, and possibly 
> Ivan,
> as I believe that if you don't they will continue to receive copies of any RT 
> issues etc?
> 
> cheers,
> Neil
> 


 End of forwarded message 

-- 
andreas

Re: Mailinglist accounts

[Andreas J. Koenig]

> On Tue, 8 Nov 2011 02:34:07 +, Matt S Trout  
> said:

  > How do PAUSE ids get added and removed from these? RJBS isn't currently on
  > the P5P list and it rather seems like he should be - and I can't figure out
  > how to make that go.

Nobody is in the P5P list. List to user mapping is implemented in the
list2user table which has no web interface. I would have to run an
insert statement for that. Now the question is, what do you mean with
"it rather seems like he should be"?

-- 
andreas

Re: Module submission User::Config

[Andreas J. Koenig]

> On Wed, 2 Nov 2011 21:36:31 +0100, "Perl Authors Upload Server" 
>  said:

  > The following module was proposed for inclusion in the Module List:

  >   modid:   User::Config
  >   DSLIP:   bdpOp
  >   description: Interface for per-user parameters
  >   userid:  BTIETZ (Benjamin Tietz)
  >   chapterid:   12 (Opt_Arg_Param_Proc)
  >   communities:

  >   similar:

Are you saying there is nothing on CPAN that deals with similar
concepts? Usually it helps to identify the right namespace or the right
approach by making a review of things one has considered as similar.

-- 
andreas

Re: Taking permission to submit new version of Curses::UI

[Andreas J. Koenig]

> On Tue, 1 Nov 2011 19:17:38 +, Dmytro Gorbunov 
>  said:

  > Hello,
  > I have question regarding new release of Cursese::UI module.

  > About 9 months ago I've submitted patch for critical error
  > https://rt.cpan.org/Public/Bug/Display.html?id=54182 , also I can see
  > some unanswered from mainteiners reported problems with provided solutions.

  > May I issue new 'official' release of this module?

Have you read 
https://pause.perl.org/pause/authenquery?ACTION=pause_04about#takeover


Thanks,
-- 
andreas

Re: New module Archive::MPQ

[Andreas J. Koenig]

> On Mon, 31 Oct 2011 23:18:49 +0200, Alex Varyanick  
> said:

  > Sorry, but I have not wait enough for your answer and already renamed
  > module to Data::MPQ :)
  > Would you please give me Data::MPQ?

Done.

Thanks,
-- 
andreas

Re: Maintaining DateTime::Calendar::Hebrew

[Andreas J. Koenig]

> On Mon, 31 Oct 2011 11:48:23 +, Raphael Mankin  
> said:

  > DateTime::Calendar::Hebrew  contains a coouple of bugs w.r.t to
  > timezones and date-time arithmetic. I have tried to contact the author,
  > but mail to him bounces.

  > Could I be added to the maintainer list or, alternatively, take over the
  > module? I need to use this module in my own DataTime::Event::Jewish.

  > My PAUSE id is RAPMANKIN.

Have you talked to the author? If not, please do, maybe with CC to
modules@perl.org or maybe in an RT ticket, whatever you feel
appropriate. More guidelines about taking over modules is at

https://pause.perl.org/pause/authenquery?ACTION=pause_04about#takeover

Thanks,
-- 
andreas

Re: Please de-register the following namespaces (fixed)

[Andreas J. Koenig]

> On Wed, 26 Oct 2011 21:25:06 +0200, "Flavio S. Glock"  
> said:

 >> Looks like v6 needs to have this in its META.yml:
 >> 
 >>    no_index:
 >>      namespace:
 >>        - Lit
 >>        - Val
 >>      package:
 >>        - Apply
 >>        - Bind
 >>        - CompUnit
 >>        - Decl
 >>        - For
 >>        - If
 >>        - Index
 >>        - Lookup
 >>        - Method
 >>        - Pair
 >>        - Sig
 >>        - Sub
 >>        - Use
 >>        - While
 >> 

  > v6-0.043 was uploaded with the "no_index" in META.yml - however, based
  > on the PAUSE indexer report, it seems that the problem persists.

The tarballs 0.043 and 0.044 contain META.yml files with the same
no_index section:

  no_index:
  directory:
  - t
  - inc

So maybe you edited your META.yml but MakeMaker overwrote it?

-- 
andreas

Re: Please de-register the following namespaces (fixed)

[Andreas J. Koenig]

> On Sat, 22 Oct 2011 10:38:54 +0200, Ævar Arnfjörð Bjarmason 
>  said:

  > so please just remove our ownership of the following
  > namespaces (usernames AVAR and FGLOCK):

[...]

The following package names are also in v6-0.042:

+-+
| Apply   | 
| Bind| 
| CompUnit| 
| Decl| 
| For | 
| If  | 
| Index   | 
| Lit::Array  | 
| Lit::Code   | 
| Lit::Hash   | 
| Lit::Object | 
| Lookup  | 
| Method  | 
| Pair| 
| Sig | 
| Sub | 
| Use | 
| Val::Bit| 
| Val::Buf| 
| Val::Int| 
| Val::Num| 
| Val::Object | 
| Val::Undef  | 
| While   | 
+-+

To avert undesired effects I leave permissions for these untouched.
Please make up your mind about v6-0.042.

I find no problem with the following 63 namespaces and delete them from
your and Flavio's permission table:

+-+
| Assign  | 
| BEGIN   | 
| COMPILER| 
| Capture | 
| Closure | 
| Code| 
| Grammar | 
| Junction| 
| KindaPerl6  | 
| KindaPerl6::Grammar | 
| KindaPerl6::Grammar::Regex  | 
| KindaPerl6::Perl5::Match| 
| KindaPerl6::Runtime::Perl5::DispatchSugar   | 
| KindaPerl6::Runtime::Perl5::DispatchSugar::Dispatch | 
| KindaPerl6::Runtime::Perl5::MOP | 
| KindaPerl6::Runtime::Perl5::Wrap| 
| KindaPerl6::Traverse| 
| KindaPerl6::Visitor::EmitAstHTML| 
| KindaPerl6::Visitor::EmitHTML   | 
| KindaPerl6::Visitor::EmitLisp   | 
| KindaPerl6::Visitor::EmitParrot | 
| KindaPerl6::Visitor::EmitPerl5  | 
| KindaPerl6::Visitor::EmitPerl5InlineToken   | 
| KindaPerl6::Visitor::EmitPerl5Regex | 
| KindaPerl6::Visitor::EmitPerl6  | 
| KindaPerl6::Visitor::EmitTokenC | 
| KindaPerl6::Visitor::ExtractRuleBlock   | 
| KindaPerl6::Visitor::Global | 
| KindaPerl6::Visitor::Hyper  | 
| KindaPerl6::Visitor::MetaClass  | 
| KindaPerl6::Visitor::Namespace  | 
| KindaPerl6::Visitor::Perl   | 
| KindaPerl6::Visitor::RegexCapture   | 
| KindaPerl6::Visitor::Scope  | 
| KindaPerl6::Visitor::Token  | 
| Lit::NamedArgument  | 
| Lit::Pair   | 
| Lit::Seq| 
| Multi   | 
| NamedArgument   | 
| Native::Buf | 
| P5Token | 
| Pad | 
| Rule::After | 
| Rule::Before| 
| Rule::Block | 
| Rule::Capture   | 
| Rule::CharClass | 
| Rule::Concat| 
| Rule::Constant  | 
| Rule::Dot   | 
| Rule::InterpolateVar| 
| Rule::NamedCapture  | 
| Rule::NegateCharClass   | 
| Rule::Or| 
| Rule::Quantifier| 
| Rule::SpecialChar   | 
| Rule::Subrule   | 
| Rule::SubruleNoCapture  | 
| Rule::Var   | 
| Signature::Item | 
| Subset  | 
| Val::Char   | 
+-+


This translated to 90 permission records that have been removed.

Let me know in case I have overlooked something.

-- 
andreas

Re: Failed index again

[Andreas J. Koenig]

> On Fri, 14 Oct 2011 11:01:23 -0500, brian d foy  
> said:

 bdf> [[ This message was both posted and mailed: see
 bdf>the "To," "Cc," and "Newsgroups" headers for details. ]]

 bdf> In article ,
 bdf> Christopher Fields  wrote:

 >> I have no idea why this is not being indexed; is it due to the module
 >> versions in the META.yml file?
 >> 
 >> Copy of META.yml and error message here:
 >> 
 >> https://gist.github.com/1284795
 >> 
 >> I had this problem before with BioPerl-DB, it seems specific to the
 >> 'provides' section, which is being autogenerated via Module::Build and
 >> './Build dist'.   Maybe the 'version' = 0?

 bdf> Why are the module versions 0? That seems really odd. Is that the
 bdf> version in the module?

I find at least two bugs involved. One is that META.yml is lying: it
reports versions to be 0 even though the modules in that package all
assign no version at all. The second bug was in the indexer that did not
treat a version of zero as a proper version number.

I've fixed the bug in the indexer and now
CJFIELDS/BioPerl-Network-1.006901.tar.gz is indexed. I cannot guarantee
that this will work well for the enduser because endusers will have
modules without version numbers installed and they (or their software)
will possibly conclude that CPAN has indexed a different version than
what they have installed.

If I were a user I would probably report the missing versions as a bug
in the BioPerl-Network package.

All that said, thanks to Christopher for finding and reporting the bug
in the indexer. Would you fancy to bring the issue also to the attention
of the Module::Build people?

 bdf> What happens if they are not zero?

Now this would be boring, everything would simply work:)

-- 
andreas

Re: PAUSE ID request (HENKP; Henk P. Penning)

[Andreas J. Koenig]

> On Fri, 30 Sep 2011 20:01:43 -0400, David Golden  said:

  > On Fri, Sep 30, 2011 at 7:48 PM, Henk P. Penning  wrote:
 >>    For debugging purposes you can always go to PAUSE directly and inspect
 >>    the RECENT files in /authors/ or /modules/. Because what Ask produces
 >>    is a made from ingredients from PAUSE and funet and what inotify then
 >>    sees on disk. Here is a glance at current state of RECENT files on
 >>    PAUSE wrt. BSORAHAN: [etc]
 >> 
 >>  When I "went to PAUSE" I noticed that I couldn't see anything relevant
 >>  pertaining to the problem at hand ; hence my request.

  > Ah.  Even I had to do some digging as I didn't remember the URLs.
  > You don't need a PAUSE ID -- he's talking about these files:

  > ftp://pause.perl.org/pub/PAUSE/

I'm sorry for the lack of clearer communication. Yes, David is right but
I also meant the "authors" and "modules" slots in the pause rsync server
which can be accessed more efficiently than with ftp:

  % rsync pause.perl.org::
  PAUSE 
  authors   
  modules   
  scripts   
  pausedata 
  pausecode 

The rrr code was born on pause in the two directories authors/ and
modules/. It was deployed separately for the two directories. The first
incarnation of the code was (and still is) based on hooks in the PAUSE
code such that every (relevent) change in the file system is reflected
in RECENT files. So when some file is uploaded, there is some hook that
cares for the necessary manipulation in some RECENT file. Likewise when
some checksum is calculated on PAUSE's behalf there is some call to a
hook that makes sure that the according RECENT file is updated. Same for
deletions, etc.

Since CPAN is more than the two directories authors/ and modules/ we
needed a different approach on the cpan master site. Instead of having
hooks we opted for an inotify based mechanism as provided by linux.

On the CPAN master site we have inotify running, on PAUSE we have hooks.
If in doubt, hooks are better than inotify but they need more human
attention. Comparing the RECENT files in PAUSE and the cpan master will
help finding out where a bug is coming from.

-- 
andreas

Re: Namespace deletion request

[Andreas J. Koenig]

> On Fri, 23 Sep 2011 13:20:42 +0900, Honma Masahiro  
> said:

  > Hi,
  > I'd like to delete 2 namespaces;

  > Data::Libra
  > WWW::YahooJapan::KanaAddress

  > I've deleted these files from CPAN.
  > Moreover, I don't have a plan to use these namespaces.

  > I'm sorry to trouble you.

No problem, you can visit
https://pause.perl.org/pause/authenquery?ACTION=edit_mod, click the
modules and in the metadata page for these modules find "Lifecycle" at
the bottom of the screen. There pull down to "Can be deleted" and click
"Update".

The actual deletion will take place much later but you can regard it as
a virtual deletion.

Best,
-- 
andreas

Re: [perl #99212] Breech of personal information on CPAN

[Andreas J. Koenig]

> On Fri, 16 Sep 2011 14:28:34 -0500, Graham Barr  said:

  > So the list of releases that need to be removed are
  > authors/id/L/LE/LEOCHARRE/FileArchiveIndexer-1.29.tar.gz
  > authors/id/L/LE/LEOCHARRE/Image-Magick-Thumbnail-PDF-1.10.tar.gz
  > authors/id/L/LE/LEOCHARRE/Image-Magick-Thumbnail-PDF-1.11.tar.gz
  > authors/id/L/LE/LEOCHARRE/Image-Magick-Thumbnail-PDF-1.12.tar.gz
  > authors/id/L/LE/LEOCHARRE/PDF-Burst-1.14.tar.gz
  > authors/id/L/LE/LEOCHARRE/PDF-GetImages-1.13.tar.gz
  > authors/id/L/LE/LEOCHARRE/PDF-GetImages-1.17.tar.gz
  > authors/id/L/LE/LEOCHARRE/PDF-OCR2-1.09.tar.gz
  > authors/id/L/LE/LEOCHARRE/PDF-OCR2-1.10.tar.gz
  > authors/id/L/LE/LEOCHARRE/PDF-OCR2-1.11.tar.gz
  > authors/id/L/LE/LEOCHARRE/PDF-OCR2-1.12.tar.gz
  > authors/id/L/LE/LEOCHARRE/PDF-OCR2-1.13.tar.gz
  > authors/id/L/LE/LEOCHARRE/PDF-OCR2-1.19.tar.gz
  > authors/id/L/LE/LEOCHARRE/PDF-OCR2-1.20.tar.gz
  > authors/id/L/LE/LEOCHARRE/Zack-1.08.tar.gz

  > these have already been removed from search.cpan.org and cpansearch.perl.org

And they are now also removed from the master cpan site pause.perl.org.

Thanks, Graham, for compiling the list, and thanks Charles Hamilton for
your patience. Sorry for the troubles. 

For a reference, the relevant parts of this thread have been stored at

http://www.xray.mpe.mpg.de/mailing-lists/modules/2011-09/msg00176.html
http://www.xray.mpe.mpg.de/mailing-lists/modules/2011-09/msg00177.html

Sorry again and hopefully this matter will be resolved between the
author and you without further involvement of the CPAN,
-- 
andreas

Re: Indexing of XML-SAX-Base

[Andreas J. Koenig]

> On Tue, 06 Sep 2011 12:37:19 +1200, Grant McLean  
> said:

  > On Mon, 2011-09-05 at 19:09 -0500, brian d foy wrote:
 >> [[ This message was both posted and mailed: see
 >> the "To," "Cc," and "Newsgroups" headers for details. ]]
 >> 
 >> In article <1315177320.2276.26.camel@hoiho>, Grant McLean
 >>  wrote:
 >> 
 >> > I wonder however if the CPAN indexer may have ignored my XML-SAX-Base
 >> > upload due to the presence of the version included in the then-current
 >> > release of XML-SAX.
 >> 
 >> I'm not sure what you're asking, but does looking at the
 >> 02packagedetails file answer your question? Whatever the index thinks
 >> is authoritative will be listed in the file.

  > Indeed - that's what prompted my question :-)

  > The most up-to-date version of that file that I can find says:

  >   XML::SAX::Base1.02  K/KH/KHAMPTON/XML-SAX-Base-1.02.tar.gz

  > Given that I uploaded version 1.05 yesterday and given that according to
  > PAUSE I have co-maint permission on the distribution, why was it not
  > indexed?

The tarball does not contain any file Base.pm, so the indexer cannot
intuit that you just uploaded XML::SAX::Base. You can convince the
indexer by writing in the META.yml a "provides" entry instead.

I'm sorry that you did not get any email about this. The indexer just
gives up when a tarball has no pm file and no provides entry. I think
this is fixable, at least can be done better. ==> TODO

-- 
andreas

Re: Fixing the mixed-content warnings on https://pause.perl.org/

[Andreas J. Koenig]

> On Thu, 17 Mar 2011 19:44:24 -0400, Daniel Kahn Gillmor 
>  said:

  > Another way would be to use a "scheme-relative" value for the src:

  >   src="//www.fiz-chemie.de/fileadmin/template/img/logos/FIZ_Logo_4c.gif" />

  > You can read more about scheme-relative URIs here:

  >  http://stackoverflow.com/questions/3583103/scheme-relative-urls

  > hope this is useful,

It is! Daniel++

Thank you very much. It's in. Unfortunately I had disabled the firefox
warning some time ago so never got reminded about the problem.

Thanks for the heads up and for the lecture on scheme-relative URIs that
I didn't know.

Best,
-- 
andreas

Re: Module submission Bundle::CommonModules

[Andreas J. Koenig]

> On Tue, 01 Mar 2011 00:37:50 -0600, brian d foy  
> said:

  > bundles are
  > ancient technology. Task::* is the way to go now. :)

Objection! Bundle:: and Task:: both have their merits. A Bundle can
determine the order in which the objects get installed. It may name
distributions directly, so can extend into unindexed territory. And a
Bundle:: updates its collected modules to highest available version
number.

While you may be right that Task:: is appropriate in this case time is
not yet ripe for Bundle:: bashing:)

-- 
andreas

Re: HTML::Formatter module maintenance

[Andreas J. Koenig]

>>>>> On Fri, 18 Feb 2011 12:05:36 +, Nigel Metheringham 
>>>>>  said:

  > On 14 Feb 2011, at 06:26, Andreas J. Koenig wrote:
 >> Nigel, I made you co-maintainer as well. Please keep trying to find Jeff
 >> and Sean because if they are both unreachable we should give the primary
 >> role of the primary maintainer to somebody else.

  > I still seem to have only have co-maint on 
  > HTML::FormatRTF
  > HTML::Formatter

  > but no rights on 
  > HTML::FormatPS
  > HTML::FormatText

  > could these 2 modules (part of the http://search.cpan.org/dist/HTML-Format/ 
set)
  > be granted to me too...

Fixed.

  > [I tried to push a release up which the PAUSE indexer got upset about.
  > Do I need to retry this once the permissions are fixed? ]

Normally, you would have to tell the indexer to reindex but I've taken
care of that.

-- 
andreas

Re: SPOON can get removed from PAUSE

[Andreas J. Koenig]

> On Mon, 14 Feb 2011 14:41:24 -0500, David Golden  said:

  > Andy,

  > I have removed you as contact for SPOON, per your request.  Thank you
  > for being the point of contact for several years.

  > Historical practice seems to be that unless a person such as yourself
  > (or brian d foy for Randy Kobes) steps up to be point of contact,
  > email is directed instead to modules@perl.org, so I have done that
  > with SPOON.

I'm tempted to turn off email forwarding. Since Andy says there's only
spam left, there's nothing to gain by giving spamassassin at perl.org
work forever. Hmm. I just did that. If there are arguments against, this
is reversible.

-- 
andreas

Re: HTML::Formatter module maintenance

[Andreas J. Koenig]

> On Sun, 13 Feb 2011 17:57:23 +0100, Gisle Aas  
> said:

  > On Feb 13, 2011, at 16:04 , Nigel Metheringham wrote:
 >> Would you be willing to support me taking co-maintainer over this?

  > I would not mind, but I don't have permission at PAUSE for changing
  > this. I'm just a co-maintainer myself.

  > It looks like JGOFF was set as primary maintainer at some point; but
  > he has not released anything; . I
  > hope we can get Andreas to override this.

 >> My PAUSE id is NIGELM
 >> 
 >> 
 >> Nigel.

Nigel, I made you co-maintainer as well. Please keep trying to find Jeff
and Sean because if they are both unreachable we should give the primary
role of the primary maintainer to somebody else.

-- 
andreas

Re: Adopting Text-Format from Gabor Egressy

[Andreas J. Koenig]

> On Thu, 27 Jan 2011 16:22:36 +0200, Shlomi Fish  
> said:

  > Now that we've established that we are unable to reach GABOR (Gabor Egressy 
- 
  > not to be confused with SZABGAB/Gabor Szabo), and he has not responded for 
two 
  > months minus a week and considering the fact that he hasn't shown any CPAN 
  > activity since 1998, may I get a comaint bit for Text-Format so I can 
update 
  > it?

Yes, of course. I'm sorry for my confusion. I've given you comaint just
now.

Gabor, if you read this, note that the module stays yours and the change
in the permissions is a reversible thing.

-- 
andreas

Re: HTML::FormatData

[Andreas J. Koenig]

> On Fri, 31 Dec 2010 19:36:14 +, Nigel Horne  
> said:

  > Is HTML::FormatData still being maintained?  It has had no work for 5
  > years even though it has a number of CPAN failures and outstanding
  > bugs.

  > If not, could I take over its maintenance?  My pause ID is NHORNE.

Thanks for your offer. There is a guideline how to proceed at

   http://www.cpan.org/modules/04pause.html#takeover

Thanks again,
-- 
andreas

Re: Failed: PAUSE indexer report TJENNESS/Astro-FITS-Header-3.02.tar.gz

[Andreas J. Koenig]

> On Fri, 31 Dec 2010 10:50:39 -0500, David Golden  said:

  > Just a thought from past version number oddities: is PAUSE a 64-bit
  > perl and is the comparison a numeric one?  It's remarkable how X !=
  > 'X' sometimes.

No, it isn't

% /usr/local/perl-5.12.0-RC5/bin/perl -V  
Summary of my perl5 (revision 5 version 12 subversion 0) configuration:
   
  Platform:
osname=linux, osvers=2.6.8-2-686-smp, archname=i686-linux
uname='linux pause 2.6.8-2-686-smp #1 smp tue aug 16 12:08:30 utc 2005 i686 
gnulinux '
config_args='-Dinstallusrbinperl=n -Uversiononly -Dusedevel -des 
-Dprefix=/usr/local/perl-5.12.0-RC5'
hint=recommended, useposix=true, d_sigaction=define
useithreads=undef, usemultiplicity=undef
useperlio=define, d_sfio=undef, uselargefiles=define, usesocks=undef
use64bitint=undef, use64bitall=undef, uselongdouble=undef
usemymalloc=n, bincompat5005=undef
  Compiler:
cc='cc', ccflags ='-fno-strict-aliasing -pipe -fstack-protector 
-I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64',
optimize='-O2',
cppflags='-fno-strict-aliasing -pipe -fstack-protector -I/usr/local/include'
ccversion='', gccversion='4.1.2 20061115 (prerelease) (Debian 4.1.1-21)', 
gccosandvers=''
intsize=4, longsize=4, ptrsize=4, doublesize=8, byteorder=1234
d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=12
ivtype='long', ivsize=4, nvtype='double', nvsize=8, Off_t='off_t', 
lseeksize=8
alignbytes=4, prototype=define
  Linker and Libraries:
ld='cc', ldflags =' -fstack-protector -L/usr/local/lib'
libpth=/usr/local/lib /lib /usr/lib
libs=-lnsl -ldb -ldl -lm -lcrypt -lutil -lc
perllibs=-lnsl -ldl -lm -lcrypt -lutil -lc
libc=/lib/libc-2.7.so, so=so, useshrplib=false, libperl=libperl.a
gnulibc_version='2.7'
  Dynamic Linking:
dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags='-Wl,-E'
cccdlflags='-fPIC', lddlflags='-shared -O2 -L/usr/local/lib 
-fstack-protector'


Characteristics of this binary (from libperl): 
  Compile-time options: PERL_DONT_CREATE_GVSV PERL_MALLOC_WRAP PERL_USE_DEVEL
USE_LARGE_FILES USE_PERLIO USE_PERL_ATOF
  Locally applied patches:
 RC5
  Built under linux
  Compiled at Apr 11 2010 18:09:13
  @INC:
/usr/local/perl-5.12.0-RC5/lib/site_perl/5.12.0/i686-linux
/usr/local/perl-5.12.0-RC5/lib/site_perl/5.12.0
/usr/local/perl-5.12.0-RC5/lib/5.12.0/i686-linux
/usr/local/perl-5.12.0-RC5/lib/5.12.0
.


-- 
andreas

Re: Adopting Text-Format from Gabor Egressy

[Andreas J. Koenig]

> On Wed, 29 Dec 2010 09:53:08 +0200, Gabor Szabo  said:

 >> No. I cannot believe that a man who has an active blog cannot be
 >> contacted. http://szabgab.com/ latest entry is 12 days old. Did you
 >> really try hard?

  > That's a different Gabor.

Oops, sorry!

  > I can ask on the Hungarian Perl Mongers list if anyone knows Gabor Egressy

Thank you!
-- 
andreas

Re: Adopting Text-Format from Gabor Egressy

[Andreas J. Koenig]

> On Tue, 28 Dec 2010 20:33:03 +0200, Shlomi Fish  
> said:

   > Hi all,
   > On Friday 10 Dec 2010 20:13:21 Shlomi Fish wrote:
 >> Hi all,
 >> 
 >> On Friday 03 December 2010 19:03:08 Shlomi Fish wrote:
 >> > Hi,
 >> > 
 >> > I'd like to adopt the Text-Format distribution from GABOR:
 >> > 
 >> > http://search.cpan.org/dist/Text-Format/
 >> > 
 >> > "GABOR" has not updated any of his modules since 1998. Gábor, if you're
 >> > there, can you approve of this request?
 >> 
 >> I've sent this E-mail 7 days ago and did not hear back from you guys. Is
 >> this request being processed? Also see what I've writen about:
 >> 
 >> http://search.cpan.org/dist/Template-Stash-AutoEscape/
 >> 

   > OK, now you can see I've sent the original E-mail more than 3 weeks ago. 
Has 
   > there been any progress?

No. I cannot believe that a man who has an active blog cannot be
contacted. http://szabgab.com/ latest entry is 12 days old. Did you
really try hard?

-- 
andreas

Re: Failed: PAUSE indexer report AYOUNG/PPIx-IndexLines-0.03.tar.gz

[Andreas J. Koenig]

> On Thu, 9 Dec 2010 19:22:09 -0700, Alan Young  
> said:

 >> The following packages (grouped by status) have been found in the distro:
 >> 
 >> Status: Permission missing
 >> ==
 >> 
 >>     module: PPI::Document
 >>    version: 0.03
 >>    in file: lib/PPIx/IndexLines.pm
 >>     status: Not indexed because permission missing. Current registered
 >>             primary maintainer is ADAMK. Hint: you can always find the
 >>             legitimate maintainer(s) on PAUSE under "View Permissions".

  > Is this a non-issue or is it not acceptable to add a method to an
  > object for a third party module?

We don't judge acceptability. It's a namespace clash and the involved
parties in such a clash are responsible to deal with it. I'd suggest you
ask on PPI channels for prior art and conventions.

-- 
andreas

Re: P5P-owned CPAN modules

[Andreas J. Koenig]

>>>>> On Thu, 9 Dec 2010 13:22:40 -0500, Jesse Vincent  said:

  > On Thu, Dec 09, 2010 at 08:19:15AM +0100, Andreas J. Koenig wrote:
 >> Sorry, I have missed what the original question was. Was it RAFL wants
 >> permissions to maintain ExtUtils::Command?

  > RAFL is working to dual life / update the cpan versions of numerous core
  > modules, a few at a time.  I'd rather make this easier, rather than
  > harder for him ;)

This is an excellent reason, of course, sorry I didn't get it. And now
looking back through the history of this thread I can explain what made
me confused. Your first message with the subject "P5P-owned CPAN
modules" had an In-Reply-To header of

  In-Reply-To: 

pointing to an unrelated posting.

All's well, move on, there's nothing to see here...

Cheers,
-- 
andreas

Re: P5P-owned CPAN modules

[Andreas J. Koenig]

> On Thu, 09 Dec 2010 07:54:12 +0900, brian d foy  
> said:

 >> Could you please grant RAFL comaint permissions on all modules owned by P5P
 >> through whatever mechanism makes most sense on the PAUSE side

Most sense makes to do what is actually needed. Like in the relation of
variables and scope, we prefer narrow scopes. Minimizing the number of
permission assignments, not maximizing. Not because we distrust anybody
but because we want to minimize bookkeeping.

Sorry, I have missed what the original question was. Was it RAFL wants
permissions to maintain ExtUtils::Command?

  > [...]
  > Added FLORA to co-maintainers of ExtUtils::Command.
  > [...]

So that might have been the answer?

-- 
andreas

Re: Resurrecting old account

[Andreas J. Koenig]

> On Tue, 30 Nov 2010 09:28:53 -0500, David Megginson 
>  said:

  > I'd prefer that to david.meggin...@megginson.com
  > if the address is going to be published?

Every communication on this mailing list is published.

I've changed your email address to david.meggin...@megginson.com now.
Don't hesitate to ask me if problems remain.

-- 
andreas

Re: Resurrecting old account

[Andreas J. Koenig]

> On Sat, 27 Nov 2010 11:11:52 -0500, David Megginson
>  said:

Hi David,

  > I have an old CPAN account from the 1990s under the name "dmegg".
  > Unfortunately, I no longer have access to whatever email account I
  > associated with that (possibly one in the uottawa.ca domain), so I
  > cannot use the automated password recovery.

Sure you can. The email address you have with your account is public and
published, so I can reproduce it here: da...@megginson.com

Looks like you can receive mail there.

  > I would like to transfer the SGMLSpm module to Andrew Bailey (RAAB),
  > who is interested in actively maintaining it.  I would appreciate any
  > help, either getting me access to the account (so that I can do the
  > transfer), or transferring ownership of the module directly.

I'm ready to help if you encounter problems but I'm pretty sure you can
use the password recovery mechanism.

-- 
andreas

Re: I think that clears the PAUSE request backlog

[Andreas J. Koenig]

> On Mon, 22 Nov 2010 07:36:39 -0500, David Golden  said:

  > On Sun, Nov 21, 2010 at 11:59 PM, brian d foy  wrote:
 >> I don't pay attention to new user requests since Johan used to do all
 >> of that before I got to it. I guess he's disappeared.

  > I volunteered to be a PAUSE admin in response to the PAUSE ID request
  > backlog.  I'm an "inbox zero" kind of guy, so and extra couple clicks
  > per day is (so far) not a problem for me.

  > I wouldn't mind seeing a recaptcha added to the request form -- other
  > than spam I'm not really sure why we'd say "no" to a request.

Here in Europe I've heard for years that it is easy to solve captchas.

http://www.networkworld.com/news/2010/111910-wiseguy-scalpers-bought-tickets-with.html

Sure, they could even more easily fool us, but probably not with
thousands of accounts.

-- 
andreas

Re: I think that clears the PAUSE request backlog

[Andreas J. Koenig]

> On Mon, 22 Nov 2010 06:59:18 +0200, brian d foy  
> said:

  > I don't pay attention to new user requests since Johan used to do all
  > of that before I got to it. I guess he's disappeared.

  > The new alpha page looks very cool, but in some of the boxes I'm
  > getting raw JSON as text instead of formatted text.

Well, it's intentional; it's canonized and indented JSON which seems
quite well readable to me.

-- 
andreas

Re: Fwd: PAUSE indexer report MMUSGROVE/File-LinkDir-1.00.tar.gz

[Andreas J. Koenig]

> On Sat, 20 Nov 2010 18:57:15 -0600, Matthew Musgrove 
>  said:

  > Can someone tell me what this means exactly?

I can reproduce with

  /usr/local/perl-5.10.1/bin/perl -e 'use YAML::XS; print 
YAML::XS::LoadFile("META.yml")'
  YAML::XS::Load Error: The problem:

  Invalid trailing UTF-8 octet

  was found at document: 0

Looks like your META.yml was written in latin1 which I think is not
legal YAML.

  > It looks like it was indexed.

Yes.

  > Do I need to fix META.yml and upload as File-LinkDir-1.01-tar-gz?

I wouldn't sweat over it. If nobody complains you can fix it next time
you do a regular upload. As a PAUSE maintaintainer I see no immediate
reason to ask you to re-upload.

  > If so,
  > how?

I don't know what you or some of the involved software did wrong. I see
only the result of a META.yml written in latin1.

  > BTW, I'm using Module::Install and I didn't see this with my previous
  > uploads.

Which previous upload are you talking about and what might have been
different in these? Note that developer releases do not get read by the
indexer, so those don't count.

-- 
andreas

Re: please release namespace Task::Digest

[Andreas J. Koenig]

> On Fri, 19 Nov 2010 19:13:09 -0800 (PST), gray  said:

  > Before I uploaded a new module, I thought I did my due diligence by
  > searching search.cpan.org for an existing module of the same name.
  > Nothing showed up for Task::Digest so that is what I used [1]. But it
  > appears that there once existed a distribution of the same name 3
  > years ago [2], but it has since been removed from CPAN. Also, the
  > author's CPAN page [3] does not indicate that he has registered the
  > namespace. If it is possible, please release the Task::Digest
  > namespace.

This all looks reasonable, so I did it:

"Made GRAY primary maintainer of Task::Digest."

CC'ing SKI. SKI, you still have co-maintainership status. If you want to
use it, please coordinate with GRAY. If you don't want it, you can give
it up via PAUSE in the menu "Change Permissions".

  > Thank you,

Thank you for your effort!

  > [1] http://search.cpan.org/dist/Task-Digest/
  > [2] http://backpan.perl.org/authors/id/S/SK/SKI/Task-Digest-0.01.meta
  > [3] http://search.cpan.org/~ski/


Cheers,
-- 
andreas

Re: I think that clears the PAUSE request backlog

[Andreas J. Koenig]

> On Fri, 12 Nov 2010 07:26:09 -0500, David Golden  said:

  > Sorry for the flurry of new user emails. I just went through
  > everything from September and early October before I got a PAUSE admin
  > bit and tried to approve everything that looked sane.  I think Kurt
  > was still approving back in August, so I hope that gets us more or
  > less current.

Thank you so much for the heroic job on this.

  > I've asked Andreas about creating a PAUSE admin page to more easily
  > see open requests so trolling through the newsgroup for old requests
  > isn't necessary in the future.

The new page is up. I've removed several hundred pron spams and invite
you to try the new page out. Maybe I've left some bug in it or didn't
think of some aspect you would like to have integrated.

-- 
andreas

Re: Module List updates?

[Andreas J Koenig]

> On Tue, 21 Oct 2003 16:55:44 +0100, Tim Bunce <[EMAIL PROTECTED]> said:

  > On Wed, Oct 15, 2003 at 01:17:33PM +0100, Steve Hay wrote:
 >> Is the Module List 
 >> (http://www.perl.com/CPAN/modules/00modlist.long.html) ever going to get 
 >> updated again, or has that been abandoned now?
 >> 
 >> It is currently monstrously out-of-date (27 Aug 2002), which seems to 
 >> prevent me from using the "Edit Module Metadata" item on PAUSE.  (I can 
 >> only edit modules that are listed in the Module List, not those that 
 >> I've added since Aug 2002.)

  > Andreas, do you have any thoughts about this? What's the story?

The publishing frequency of the modulelist itself is not
enabling/preventing editing of metadata on PAUSE. Authors can edit
metadata as soon as a namespace proposal is confirmed by an admin.

That said, there is indeed a painful delay in the production of the
modulelist. The software I have written to produce new module lists is
very broken and I'd need a whole lot of consecutive days to redesign
it so that the module list becomes as automatic as it should be. Tuit
shortage:-(

-- 
andreas

Re: Module submission Arary::Heap2

[Andreas J Koenig]

> On Fri, 3 Oct 2003 23:03:57 +0200, <[EMAIL PROTECTED] ( Marc) (A.) (Lehmann )> 
> said:

  > On Tue, Sep 23, 2003 at 12:56:58PM +0800, Autrijus Tang <[EMAIL PROTECTED]> wrote:
 >> ??? ?|, 2003-09-18 15:47, Perl Authors Upload Server ? 
 >> > The following module was proposed for inclusion in the Module List:
 >> >   modid:   Arary::Heap2
 >> 
 >> "Arary" will not do.  Please fix this typo and re-submit. :-)

 >> > And here is the rationale for choosing Array::Heap2:
 >> > Well, there is no good rationale, but while registering, I found
 >> > that Array::Heap has been registered in what seems 1998 already,
 >> > however, a module hasn't been uploaded in the last 5 years. (I
 >> > didn'T see this at first because my searches were for existing
 >> > modules only :)
 >> > I've contacted the author, but haven't received a reply yet (but
 >> > it's not been a long time).
 >> 
 >> Interesting.  CC'ed to John Macdonald, to see whether he is willing to
 >> abdicate that namespace to you.

  > So far, no reaction (but that could just mean he's on vacation or so).
  > Anyways, since it seems that he didn't even register that namespace
  > himself, I'll CC: the person responsible for the registration, Andreas
  > König :) Maybe he remembers what happened back then in more detail. Maybe
  > not...

  > I can't come up with the reference (despite having found it quickly the
  > first time, but now all my google searches fail to find the mail where
  > andreas says he'll register that modulelist entry).

  > Ah well :)

Google doesn't index printed books. Mastering Algorithms with Perl
promises on page 101:

the routines to apply binary heap ordering to a user-provided
array will be put in a separate module called Array::Heap.

I put the two co-authors on the CC. Maybe they know more about the
fate of said module.


-- 
andreas

Re: Uploaded wrong file to PAUSE by mistake

[Andreas J Koenig]

> On Sat, 20 Sep 2003 15:28:55 +0200, "Philip Newton" <[EMAIL PROTECTED]> said:

  > Hello, I just wanted to upload two of my modules to CPAN. For one of 
  > them, I accidentally clicked on someone else's module instead.

  > That module is "Log-Log4perl-0.37.tar.gz [161253b; MSCHILLI, PNE]"

There was enough time between his upload and your click, so he already
owned it when you clicked. So no harm done.

  > It should not belong to me (presumably it should belong to MSCHILLI 
  > only). Can you fix this for me, please? Do I have to do something? (For 
  > example, delete it from my CPAN directory?)

Yes, please delete it, nothing else is needed.

  > Apologies.

NO problem . . . Best Regards,
-- 
andreas

Re: Taking Over CGI::Lite

[Andreas J Koenig]

> On Sun, 10 Aug 2003 18:55:25 +0100, Smylers <[EMAIL PROTECTED]> said:

  > So I'd like to take over CGI::Lite.  If you (and [EMAIL PROTECTED]) are
  > happy with this then I'll fill in the form to get a CPAN account.

Yes, please!

-- 
andreas

Re: PAUSE indexer report VKON/Tcl-0.7.tar.gz

[Andreas J Koenig]

"Vadim Konovalov" <[EMAIL PROTECTED]> writes:

> Dear [EMAIL PROTECTED] Dear Andreas!
>
> Could you please help me becoming a co-maintainer of namespaces Tcl and
> Tcl::Tk?
> Should I perform some steps in order to do that?
>
> Previous maintainer, Malcolm Beattie seems to be not reachable right now.
> I tried to contact him and I have not succeeded.
>
> Besides, I sucessfully released few beta releases after that and I hope
> those modules could be quite useful for some people (including me of
> course). Also I am planning to do more releases of those.

Given that Malcolm hasn't uploaded anything for 2 1/2 years and that I
haven't heard anything from him for about that time and that you
already have uploaded 4 well respected releases for the two namespaces
(Thanks!), I'd say it is justified to make you co-maint without any
further action and delay. I'm pretty sure, Malcolm will welcome this
decision.

Done! You can either schedule re-indexing or uploaded a fresh package
to see the change in effect.

Regards,
-- 
andreas

Re: pause.perl.org down all day?

[Andreas J Koenig]

"Soren Andersen" <[EMAIL PROTECTED]> writes:

> Hi, Andreas,
>
> Since last November I haven't uploaded to PAUSE and today I am finally
> trying to get the fixed release of my Filesys::CygwinPaths (0.04)
> uploaded. Unless I am doing something incorrectly, the pause.perl.org is
> completely offline. Can you please confirm?
>
> I've been trying for about 5 hours here on US East Coast. The time now is
> about 7:10 PM local (ca. 19:10 EDT).

I can confirm, we have a yet unidentified hardware problem. Please
retry from time to time with patience.

-- 
andreas

Re: PAUSE ID request (KAFKA; Nicolas Herry)

[Andreas J. Koenig]

> On Sat, 05 Apr 2003 20:51:21 -0900, "Sean M. Burke" <[EMAIL PROTECTED]> said:

  > At 10:39 PM 2003-04-05 +0200, Johan Vromans wrote:
 >> "Sean M. Burke" <[EMAIL PROTECTED]> writes:
 >> > Why are these email addresses censored?
 >> To avoid them being exposed on the mailing list.
  > That is perilously close to a useless tautology.

PAUSE promises that there is a secret email address that is never
exposed to third parties. Only PAUSE and its admins should see it. The
mailing list is archived for everybody, so we need to be careful with
what we display on the list.

For the same reason I urge all admins to always use the
[EMAIL PROTECTED] mail address when communicating with authors.

-- 
andreas

Re: PAUSE indexer report ELIZABETH/forks-0.03.tar.gz

[Andreas J. Koenig]

> On Wed, 2 Apr 2003 12:02:34 +0200, Elizabeth Mattijsen <[EMAIL PROTECTED]> said:

 >> It means that the PAUSE indexer does not pick up the namespaces
 >> "threads" and "threads::shared" that are contained in your files
 >> lib/forks.pm and lib/forks/shared.pm respectively.

  > Ok, does this have to do with search.cpan.org?

Not directly. The namespace index on PAUSE is
CPAN/modules/02packages.details.txt.gz. I don't know how exactly this
index influences search.cpan.org.

 >> If you visit the
 >> "View Permission" page on PAUSE, you find out that these two
 >> namespaces are co-owned by Artur and Jarkko.
 >> If this is what you want, all is OK, you need not react to this email.

  > Well, I don't want to update the "real" threads.pm and
  > threads::shared.pm, but I would like people to find forks.pm if
  > they're looking for threads.pm

You mean, if people ask CPAN.pm to update threads, they should
actually get forks.pm? If you want that, you need to ask Artur and
Jarkko if they make you co-maintainer. To me this looks dangerous.

The PAUSE indexer has a 1 to 1 mapping of namespaces to distributions,
i.e. only one file on CPAN can be mapped to a given namespace.

search.cpan.org is a different beast. It can list unlimited documents
associated with a given query term.

 >> Please let me know how the wording of the notification could be
 >> improved to avoid confusion.

  > I'm still confused as to what "indexed" means in this context, so I
  > can't suggest an alternate wording (yet)...

  % zcat 02packages.details.txt.gz | head
  File: 02packages.details.txt
  URL:  http://www.perl.com/CPAN/modules/02packages.details.txt
  Description:  Package names found in directory $CPAN/authors/id/
  Columns:  package name, version, path
  Intended-For: Automated fetch routines, namespace documentation.
  Written-By:   Id: mldistwatch 330 2003-03-31 12:22:25Z k 
  Line-Count:   17653
  Last-Updated: Wed, 02 Apr 2003 09:41:32 GMT

  A::B  undef  M/MA/MARCEL/Devel-SearchINC-0.01.tar.gz

 >> > And what should I put in a META.yml file to prevent this problem
 >> > (if at all possible?).
 >> META.yml support is currently in preparation. I don't know when it
 >> will start to influence the indexer.

  > Ok, maybe then it should not say that it didn't find a META.yml file,
  > as that _is_ confusing at the current time...

Fair enough. Fixed.

-- 
andreas

Re: PAUSE indexer report ELIZABETH/forks-0.03.tar.gz

[Andreas J. Koenig]

> On Wed, 2 Apr 2003 11:17:07 +0200, Elizabeth Mattijsen <[EMAIL PROTECTED]> said:

  > At 11:08 +0200 4/2/03, PAUSE wrote:
 >> The following report has been written by the PAUSE namespace indexer.
 >> Please contact [EMAIL PROTECTED] if there are any open questions.
 >> Id: mldistwatch 330 2003-03-31 12:22:25Z k
 >> 
 >> User: ELIZABETH (Elizabeth Mattijsen)
 >> Distribution file: forks-0.03.tar.gz
 >> Number of files: 16
 >> *.pm files: 6
 >> README: forks-0.03/README
 >> META.yml: No META.yml found
 >> 
 >> Timestamp of file: Wed Apr  2 09:03:41 2003 UTC
 >> Time of this run: Wed Apr  2 09:08:28 2003 UTC
 >> 
 >> The following packages have been found in the distro:
 >> 
 >> threads
 >> version: 0.03
 >> in file: forks-0.03/lib/forks.pm
 >> status: Not indexed because permissions missing. Visit PAUSE and
 >> click "View Permissions" to find the legitimate
 >> maintainer(s).
 >> 
 >> threads::shared
 >> version: 0.03
 >> in file: forks-0.03/lib/forks/shared.pm
 >> status: Not indexed because permissions missing. Visit PAUSE and
 >> click "View Permissions" to find the legitimate
 >> maintainer(s).

  > Interesting problem with forks.pm.  I'm not sure what "indexed" means
  > in this respect.

It means that the PAUSE indexer does not pick up the namespaces
"threads" and "threads::shared" that are contained in your files
lib/forks.pm and lib/forks/shared.pm respectively. If you visit the
"View Permission" page on PAUSE, you find out that these two
namespaces are co-owned by Artur and Jarkko.

If this is what you want, all is OK, you need not react to this email.

Please let me know how the wording of the notification could be
improved to avoid confusion.

  > And what should I put in a META.yml file to prevent this problem
  > (if at all possible?).

META.yml support is currently in preparation. I don't know when it
will start to influence the indexer.

-- 
andreas

Re: Welcome new user PMAREK

[Andreas J. Koenig]

> On Mon, 31 Mar 2003 17:44:26 +0200, "Perl Authors Upload Server" <[EMAIL 
> PROTECTED]> said:

  > Welcome Philipp Marek,

  > PAUSE, the Perl Authors Upload Server, has a userid for you:

  > PMAREK

All mails to pmarek so far are bouncing. If anybody knows a working
email address of pmarek, please let him know.

Our message board isn't of any use for somebody who doesn't get the
password transmitted:-/

-- 
andreas

Re: PAUSE indexer report KWILLIAMS/Module-Build-0.17.tar.gz

[Andreas J. Koenig]

> On Sat, 29 Mar 2003 19:13:09 -0600, Ken Williams <[EMAIL PROTECTED]> said:

  > On Saturday, March 29, 2003, at 06:40  PM, PAUSE wrote:
 >> User: KWILLIAMS (Ken Williams)
 >> Distribution file: Module-Build-0.17.tar.gz
 >> Number of files: 38
 >> *.pm files: 17
 >> README: Module-Build-0.17/README
 >> META.yml: 2 META.yml found, ignoring

  > Hi,

  > The two META.yml files found are META.yml and t/Sample/META.yml in
  > this case.  I think probably the indexer should only look for the
  > META.yml file at the root level of the distribution, or perhaps ignore
  > any in t/ .  What would be best?

Fixed in revision 330. I opted for ignoring META.yml below t/ for now.


-- 
andreas

Re: Module submission again

[Andreas J. Koenig]

> On Mon, 31 Mar 2003 01:47:25 +0800, Autrijus Tang <[EMAIL PROTECTED]> said:

  > On Sun, Mar 30, 2003 at 07:22:48PM +0200, [EMAIL PROTECTED] wrote:
 >> In my opinion, again is more perlish. Especially because of its
 >> pragma-like name (compare C, C, C).

  > It is also confusing:

  > use Module::Name;
  > use again 'Module::Name';

  > If I know nothing about your module, I'd expect line two to reload
  > Module::Name unconditionally.  But maybe it's just me.  Arthur?

FWIW, I read like Autrijus.



-- 
andreas

Re: Uri update for K/KW/KWILLIAMS/AI-Categorizer-0.05.tar.gz

[Andreas J. Koenig]

> On Sat, 29 Mar 2003 00:41:32 -0600, Ken Williams <[EMAIL PROTECTED]> said:

 >> Something about the way I'm FTP-ing these files (using Net::FTP, via
 >> the 'release' tool of PETDANCE) is making them appear on PAUSE with
 >> zero length.  Not sure what that's about either.  Some Net::FTP
 >> debugging info follows:
 >> 
 >> Net::FTP=GLOB(0x38afe8)<<< 250 CWD command successful.
 >> Net::FTP=GLOB(0x38afe8)>>> PORT 192,168,0,101,213,215
 >> Net::FTP=GLOB(0x38afe8)<<< 200 PORT command successful
 >> Net::FTP=GLOB(0x38afe8)>>> STOR AI-Categorizer-0.05.tar.gz
 >> Net::FTP=GLOB(0x38afe8)<<< 421 No Transfer Timeout (60 seconds):
 >> closing control connection.
 >> Net::FTP=GLOB(0x38afe8)>>> QUIT
 >> Net::FTP: Unexpected EOF on command channel at
 >> /Library/Perl/Module/Release.pm line 323
 >> PAUSE upload successful

Maybe you're behind a firewall and need to set passive ftp?


-- 
andreas

Re: Module submission Simulation::Particle

[Andreas J. Koenig]

> On Wed, 26 Mar 2003 23:34:50 +0800, Autrijus Tang <[EMAIL PROTECTED]> said:

  > Gawd, this is embarrasing.  I made comments based on an outdated module
  > list from www.cpan.org:

  > % HEAD http://www.cpan.org/modules/00modlist.long.html |grep Modified
  > Last-Modified: Tue, 27 Aug 2002 23:40:25 GMT

It's my fault and it is indeed embarrassing. I hope that I can fix my
software that produces the module list Really Soon.

For now I can recommend you to use

perl -MCPAN::Admin -e shell

instead of the module list itself. CPAN::Admin does two things more
than CPAN.pm:

-- it displays already registered module names in green
   (it uses Term::ANSIColor)

-- it has a quite intelligent "register" command that untars a tarball
   and uses some heuristics to construct a registration request and
   finally opens a netscape/mozilla window to display the request.
   With a single click you can submit the registration.

It's not a revolution, but it has some utility.


-- 
andreas

Re: This is getting ridiculous. . .

[Andreas J. Koenig]

> On Tue, 25 Mar 2003 10:37:53 +0100, Arthur Bergman <[EMAIL PROTECTED]> said:

 >> The flow would look like this:
 >> 
 >> 1. CPAN authors use a form on PAUSE...
 >> 2. Email gets sent to RT and placed in the modules queue
 >> 3. RT sends a copy of the request to [EMAIL PROTECTED]
 >> 4. RT tracks any discussion that happens on [EMAIL PROTECTED]
 >> 5. When it is approved/rejected/altered/etc, one of the modules@
 >> admins goes into the RT system and marks it as such.
 >> 
 >> 6. Status reports can be generated that show the status of the
 >> queue.
 >> (5 in queue, 3 under discussion, 2 new)
 >> 
 >> We would be doubleplushappy to host this on the perl.org RT instance.

  > I think this would be an excellent idea.

Yes! And I'd be doubleplushappy if I could deal with RT as a (mostly)
black box.

The one problem I still see in this scenario, would be the RT ticket
number. Somewhere between (3) and (5), an admin can click on the URL
that comes with the mail/note generated by (2). By clicking on that
URL they approve the module. I'm happy that we have that
one-click-approval thingy working. But PAUSE should then be able to
tell RT about the successful approval. To do that, it seems, that
PAUSE would need the ticket number assigned by RT.

Did I express myself clearly? I hope there's a way for PAUSE to find
out the ticket number. One possible solution would run like that
(point 2 and 4a are different from above):

  1. CPAN authors use a form on PAUSE...
  2. A HTTP request is sent to RT and placed in the modules queue.
 PAUSE records the ticket number it gets from RT.
  3. RT sends a copy of the request to [EMAIL PROTECTED]
  4. RT tracks any discussion that happens on [EMAIL PROTECTED]
  4a IF an admin decides to approve by clicking on the approval
 button, PAUSE informs RT that the ticket can be closed.
  5. When it is approved/rejected/altered/etc, one of the modules@
 admins goes into the RT system and marks it as such.
  
Actually, (5) would then quite often be done automatically. Is this
doable? And if so, please let's start coding.

-- 
andreas

Re: [upload@pause.x.perl.org: Welcome new user EIMPAN]

[Andreas J. Koenig]

>>>>> On Sun, 23 Mar 2003 06:14:08 +0100, Philip Newton <[EMAIL PROTECTED]> said:

  > On Sat, 22 Mar 2003 19:14:21 +0100, [EMAIL PROTECTED] (Andreas J.
  > Koenig) wrote:

 >> I also relaxed the requirement for a valid username to /^[A-Z]{3,9}$/,

  > I suppose you mean /^[A-Z-]{3,9}$/ ? Hyphens are allowed in CPAN
  > usernames. (Though IIRC not as first or last character and not if you
  > have two consecutive hyphens.)

Historically we allowed these IDs that do not match above regex:

CHANG-LIU
IBMTORDB2
IX
JV
NI-S
P5P
PERL4LIB
RA
WIN32

In contrast, currently /^[A-Z]{3,9}$/ is hardcoded into the
registration code.

So something like /^[A-Z][-A-Z]*[A-Z0-9]+$/ does match all existing
IDs, but then the regex is missing the size restriction, which indeed
is important. In principle we do not want to be restrictive but we
also do not want an overly complicated regex. On balance, I think
/^[A-Z]{3,9}$/ is a good compromise for future usernames, but I'm open
to suggestions what the rule should be.

-- 
andreas

Re: [upload@pause.x.perl.org: Welcome new user EIMPAN]

[Andreas J. Koenig]

> On Fri, 21 Mar 2003 20:38:36 +0100, Ivo Marino <[EMAIL PROTECTED]> said:

  > Hello Andreas,
  > First of all thanks for the Perl CPAN account which the automatic
  > accounting system has created for me.

  > I would like to ask if it's possible to rename my account userid from
  > EIMPAN to simply EIM which I use by default for all the projects I'm in.

It's extremely error prone to fiddle with the innards of the database,
I would never do that. But instead I created a new account for you and
set the other one to rest.

There was no code yet that supported retiring of accounts. I have
added a checkbox that allows users to mark their accounts as ready for
retiring (but only for users that have never uploaded anything; others
can not retire).

I also relaxed the requirement for a valid username to /^[A-Z]{3,9}$/,
otherwise it wouldn't have been possible to create an account EIM.

  > Thanks in advantage for your support and thanks again for the CPAN
  > account, I really appreciate your work, continue so.

You're welcome.

-- 
andreas

Re: This is getting ridiculous. . .

[Andreas J. Koenig]

Tim, on this mailing list RT was mentioned as an idea, but I don't
know if anybody has taken steps ahead.

Robert, has anybody talked with you if and how RT could work for
managing module namespace requests? You probably know how the process
currently is structured:

1. CPAN authors can use a form on PAUSE to submit a namespace request.

2. The form generates an email to [EMAIL PROTECTED] It contains a URL
   that can be used to register the module officially.

Apart from that, PAUSE does not keep track of registrations, open
requests, altered requests, withdrawn requests, etc. Seems like an
ideal job for RT, so I ask you, if you can suggest what we need to do
to get help from RT.

Thanks for any insight you can provide,
-- 
andreas

Re: This is getting ridiculous. . .

[Andreas J. Koenig]

> On Thu, 20 Mar 2003 09:20:04 -0900 (AKST), Arthur Corliss <[EMAIL PROTECTED]> 
> said:

  > Ah, but do you really want to create a whole new top-level namespace just for
  > me?

If nobody finds an appropriate existing one, yes.

  > Keep in mind that no ^Widget:: space is currently in existence on CPAN,
  > and I thought that top-level additions were heavily discouraged.

Discouraged yes, but they are the only solution if nobody finds
anything better.

  > If you want me to resubmit under that, consider it done.

Done.

-- 
andreas

Re: This is getting ridiculous. . .

[Andreas J. Koenig]

> On Wed, 19 Mar 2003 08:30:32 -0900 (AKST), Arthur Corliss <[EMAIL PROTECTED]> 
> said:

  > Believe me, I am doing so, and I'll end up putting this under Curses::Widgets
  > if it doesn't get accepted by you.  As I mentioned above, though, *any*
  > widget, curses or otherwise, that uses fixed width fonts could benefit from
  > this, there's nothing specific in it that pertains to my somewhat warped vision
  > of a widgets framework.  I suspect that if I put it under my widget namespace,
  > though, no one using another widget framework will look at it twice, since
  > that namespace implies adoption of the framework as a whole.

Thanks for the very concise description what this module does. What
about

   Widget::FixedWidthFont::Input

or something like that. See, how I pull that name out of your
description: if something's still missing to make the name really
descriptive, please extend on that.

  [...]

  > You have been very helpful on those cases, and I appreciate that greatly.  In
  > fact, your approval on the Parse::PlainConfig took me by surprise since I
  > thought I had indicated that I would just move the active development into the
  > Config::* namespace as per brian's suggestion.

You indicated that a bit vaguely, so I went ahead. You can always
discard that namespace by changing the field "Lifecycle" in the PAUSE
interface "Edit Module Metadata".

-- 
andreas

Re: This is getting ridiculous. . .

[Andreas J. Koenig]

> On Tue, 18 Mar 2003 12:13:35 -0900 (AKST), Arthur Corliss <[EMAIL PROTECTED]> 
> said:

  > As I've said in previous conversations on the list, I'm not going to just
  > upload whatever modules I feel like, I'll only pollute the namespace I'm
  > granted by you guys.

This is the central mistake, we do not speak ex cathedra and won't and cannot.

 >> Again, I apologize that we've left you disappointed.

  > Look, I'm a big fan of CPAN and Perl in general, and I appreciate the
  > opportunities I've been given to give something back in other modules.  I just
  > want this system to work.

Thank You!

  > All I expect is this:  either "register" the module so it shows in the pause
  > interface, or tell me no, you don't think it's a worth-while addition to CPAN.

No doubt is it a useful addition to CPAN. So far you have got a single
answer that says, Text::InputRepository doesn't fly. It would be good
to hear more suggestions, not necessarily from us. How are your users
communicating? Should there not be a mailing list for your users where
you could ask for namespace suggestions? Quite often the best ideas
come from the user community. Couldn't you ask them?

  > And take a few weeks to make up your mind, if need be.  But don't leave me
  > hanging indefinitely when I have other modules that need to know what module
  > to use (i.e., is it going to be bundled with my other modules in a
  > non-polluting namespace, or will it be available on CPAN).

Maurice Makaay has taken the route to start a Curses::UI namespace
where he puts all of his modules. Wouldn't it be a good idea if you'd
do likewise?

  > If those expectations can't be met (and I don't think they're unreasonable),
  > let's find a way to recruit the community to help out.  As always, I'm more
  > than happy to help in any way possible.  But you need to give me a chance.

The crux with having a team that gives advice on namespaces is that
you cannot improve the advice by increasing the team size.

  > I'm sorry, but the last three modules have been a bear to get onto the pause
  > interface where I can more properly manage them (updates to meta-data, etc.),
  > even though I've been following the pause guidelines faithfully.

Yes, you were extraordinary unlucky so far, I'm sorry for that. I've
already analysed the case of Parse::PlainConfig in a recent posting.
Now it's about Text::InputRepository which has already been called an
unlucky choice.

I still have the impression that you're working on something that can
be called a framework. Frameworks deserve their own namespace, like
e.g. Curses::UI, Mason, Alzabo, MHonArc, POE, Zoidberg and what not.


-- 
andreas

Re: PAUSE Indexer report M/MT/MTHURN/WWW-Search-AltaVista-2.08.tar.gz

[Andreas J. Koenig]

> On Mon, 17 Mar 2003 08:02:30 -0500, "Martin 'Kingpin' Thurn" <[EMAIL PROTECTED]> 
> said:

  > Thank you Andreas,
  >   This looks good for now.  (We might have to reset WWW::Search::Scraper to
  > I-forget-his-name...)

Glenn Wood owns zillions of WWW::Search::Scraper::* modules. I "gave
up" WWW::Search::Scraper in your name, so it is floating freely again
and as soon as Glenn uploads his next version, he will get it on the
first-come-first-serve basis.

-- 
andreas

Re: templates for docs & modules?

[Andreas J. Koenig]

> On Sun, 16 Mar 2003 11:40:43 -0800, Golda Velez <[EMAIL PROTECTED]> said:

  > Hello modules folks
  > Sorry if this is a dumb question - I have been using CPAN modules for
  > years and have some of my own stuff I'd like to contribute - but I
  > haven't seen an obvious link for how to nicely format both the code
  > and docs to the CPAN 'way'. Are there templates somewhere, am I
  > missing something obvious?

perl 5.8 comes with the manpages perlmodstyle and perlnewmod. I think
they are a good starting point.

-- 
andreas

Re: PAUSE indexer report KERW/Apache-MiniWiki-0.81.tar.gz

[Andreas J. Koenig]

> On Sun, 16 Mar 2003 18:37:12 +0100, Jonas Oberg <[EMAIL PROTECTED]> said:

  > I'd be gratefule for hints how the interface failed to explain itself
  > and how it could be improved to be easier to use.

  > We did indeed find the correct buttons to press. However, when I
  > pressed the button to submit the form, nothing happened (I was
  > returned to the same page without any notice if it were successful or
  > not).

I have an idea what happened. You probably saw Apache::MiniWiki in the
selectbox and thought it was already selected, but it wasn't. I just
fixed PAUSE to give appropriate feedback, something like: "you need to
select a namespace".

Especially selectboxes with only one option to select look confusing.
So I just fixed PAUSE so that scrolling lists, that only have one
option, have this option preselected for the user.

Thanks for leading me to these buglets,
-- 
andreas

Re: PAUSE indexer report KERW/Apache-MiniWiki-0.81.tar.gz

[Andreas J. Koenig]

> On Sat, 15 Mar 2003 15:03:16 -0800, "Wim Kerkhoff" <[EMAIL PROTECTED]> said:

  > What's the trick to change the official maintainer from JONASO to KERW?
  > Jonas tried, but PAUSE didn't let him transfer permissions over to me.

Visit PAUSE, click "Change Permissions", click "2.1. Pass
maintainership status to somebody else", select "Apache::MiniWiki" and
enter "KERW" as the userid, click "Pass Maintainership Status". Done.

I've done that for you and scheduled 0.81 for re-indexing.

I'd be gratefule for hints how the interface failed to explain itself
and how it could be improved to be easier to use.

-- 
andreas

Re: PAUSE indexer report GSLONDON/Symbol-Table-1.01.tar.gz

[Andreas J. Koenig]

> On Fri, 14 Mar 2003 10:10:26 -0800 (PST), [EMAIL PROTECTED] said:

  > Hello,
  > I'm trying to upload a module to CPAN called Symbol::Table
  > I searched from the various CPAN search engines
  > to make sure I didn't bump into anyone else's namespace.
  > But when I uploaded the module, I got the following email 
  > (at bottom).

  > The email says the Symbol::Table is registered to GARROW.
  > There is nothing of that name at that CPAN account:
  > http://www.cpan.org/modules/by-authors/id/G/GA/GARROW/

  > Can I register Symbol::Table and Symbol::Table::Tie ??
  > Do I need to rename the module?

Symbol::Table::Tie is yours anyway. As we have not heard from the
owner of Symbol::Table, John Redford, for years, I've passed the
namespace to you.

Thanks!
-- 
andreas

Re: PAUSE Indexer report M/MT/MTHURN/WWW-Search-AltaVista-2.08.tar.gz

[Andreas J. Koenig]

> On Fri, 14 Mar 2003 08:32:46 -0800, John Heidemann <[EMAIL PROTECTED]> said:

  > On Fri, 14 Mar 2003 08:15:18 EST, "Martin 'Kingpin' Thurn" wrote: 
 >> Hello,
 >> many of them are registered to JOHNH as first-come.  Could you change all
 >> the WWW::Search modules to co-maint status for JOHNH and MTHURN?  Thank you.
 >> Here is a link which shows the modules that need to be changed:
 >> 
 >> http://pause.perl.org/pause/authenquery?pause99_peek_perms_by=a&pause99_peek
 >> _perms_query=JOHNH&pause99_peek_perms_sub=1

  > Please do this for all the WWW::Search modules.  Martin has been
  > maintainer for several years now.

What I did makes Martin primary maintainer for all 32 packages that
John was registered as maintainer. This is easier for me than making
co-maintainers. Please let me know if this has undesired effects. I
think, it's what you need.

mysql> select package from primeur where primeur.package like 'WWW::Search::%' a
nd primeur.userid = 'JOHNH' ;
+--+
| package  |
+--+
| WWW::Search::AltaVista   |
| WWW::Search::AltaVista::AdvancedNews |
| WWW::Search::AltaVista::AdvancedWeb  |
| WWW::Search::AltaVista::News |
| WWW::Search::AltaVista::Web  |
| WWW::Search::Crawler |
| WWW::Search::Dejanews|
| WWW::Search::ExciteForWebServers |
| WWW::Search::Fireball|
| WWW::Search::FolioViews  |
| WWW::Search::Gopher  |
| WWW::Search::HotFiles|
| WWW::Search::Infoseek|
| WWW::Search::Infoseek::Companies |
| WWW::Search::Infoseek::Email |
| WWW::Search::Infoseek::News  |
| WWW::Search::Infoseek::Web   |
| WWW::Search::Livelink|
| WWW::Search::MSIndexServer   |
| WWW::Search::Metapedia   |
| WWW::Search::NorthernLight   |
| WWW::Search::Null|
| WWW::Search::OpenDirectory   |
| WWW::Search::PLweb   |
| WWW::Search::Profusion   |
| WWW::Search::SFgate  |
| WWW::Search::Search97|
| WWW::Search::Simple  |
| WWW::Search::Snap|
| WWW::Search::Verity  |
| WWW::Search::WebCrawler  |
| WWW::Search::ZDNet   |
+--+
32 rows in set (0.00 sec)

mysql> update  primeur set userid='MTHURN' where primeur.package like 'WWW::Sear
ch::%' and primeur.userid = 'JOHNH' ;
Query OK, 32 rows affected (0.00 sec)
Rows matched: 32  Changed: 32  Warnings: 0





-- 
andreas

Re: New flag in DSLIP entry for abandoned modules?

[Andreas J. Koenig]

>>>>> On Tue, 4 Mar 2003 16:57:05 +, Tim Bunce <[EMAIL PROTECTED]> said:

  > On Tue, Mar 04, 2003 at 03:19:39PM +0100, Andreas J. Koenig wrote:
 >> Many suggestions have come over the years that we should somehow flag
 >> abandoned modules as such. Finally I came up with the simplest
 >> possible idea:
 >> 
 >> We have already the flags
 >> 
 >> S - Support Level:
 >> m   - Mailing-list
 >> d   - Developer
 >> u   - Usenet newsgroup comp.lang.perl.modules
 >> n   - None known, try comp.lang.perl.modules
 >> 
 >> I think we should add a flag "0" (zero) or "-" (minus), that has the
 >> explicit meaning: abandoned, volunteers welcome to take over maintainance.
 >> 
 >> What do you think?

  > Excellent, simple and a very good fit.

  > As for the character to use, 0 can look like O, and - could imply unknown.
  > I'd suggest "z" for zero, but that may be too 'cute', or "a" for abandoned.

I agree that 'z' looks too cute, so 'a' is it. Thanks, Tim!

-- 
andreas

Re: perl design patterns documentation only module - thoughts?

[Andreas J. Koenig]

> On Wed, 5 Mar 2003 06:19:05 -0800, Scott Walters <[EMAIL PROTECTED]> said:

  > Hi.
  > Before submitting a formal proposal, I'd like to test the waters: would there
  > be any desire for a (rather large) documentation-only module in CPAN's indexes?

  > http://wiki.slowass.net/assemble.cgi?PerlDesignPatterns

  > I've packaged that up as a module as well: 

  > 
http://www.cpan.org/modules/by-authors/id/S/SW/SWALTERS/Object-PerlDesignPatterns-0.01.tar.gz

  > Thanks very much for your time and attention,

Very nice, thanks for your work on this, it looks promising.

I think a regular upload of this to CPAN will be OK, but let's
consider our options about *how* to upload it. Looking at the sizes

  612352  PerlDesignPatterns.html
  515881  PerlDesignPatterns.pm
  420174  SWALTERS/Object-PerlDesignPatterns-0.01.tar.gz
  575971  SWALTERS/Perl_Design_Patterns_book_20021008.html

we see, the .tar.gz file is not much smaller than the HTML file.
Nobody will be interested in old versions of it. Looks like you should
just upload the PerlDesignPatterns.html regularly (maybe once a
month). Note: files matching /\.html$/ can be uploaded again and
again.

If you want to have "PerlDesignPatterns" in the modules' index, you
should probably write a module using LWP that fetches the
PerlDesignPatterns.html from the CPAN or from your server or gives
instructions how to do that. The downside would be that C< perldoc
Object::PerlDesignPatterns > will only show the manpage for the
module, but that's probably better than showing the whole document as
a manpage because it's not suited for man(1) and best viewed with a
browser.


-- 
andreas

New flag in DSLIP entry for abandoned modules?

[Andreas J. Koenig]

Many suggestions have come over the years that we should somehow flag
abandoned modules as such. Finally I came up with the simplest
possible idea:

We have already the flags

  S - Support Level:
m   - Mailing-list
d   - Developer
u   - Usenet newsgroup comp.lang.perl.modules
n   - None known, try comp.lang.perl.modules

I think we should add a flag "0" (zero) or "-" (minus), that has the
explicit meaning: abandoned, volunteers welcome to take over maintainance.

What do you think?

Background: we apparently have already users who redirect mail to
their secret and public addresses to /dev/null. I'd like to give these
people a way to abandon their modules without deleting them and
without having to talk with anybody ever again.


-- 
andreas

Re: New feature on PAUSE - messages to authors without email

[Andreas J. Koenig]

> On Tue, 25 Feb 2003 15:44:15 +0100, Johan Vromans <[EMAIL PROTECTED]> said:

  > Good job!

Thanks :-)

-- 
andreas

New feature on PAUSE - messages to authors without email

[Andreas J. Koenig]

Since today (rev. 241) PAUSE has a new feature:

Admins can put messages for individual users into the database and
these users will *always* have the message(s) on all pages they visit
near the top of the page.

This message board is a convenient way to contact users that cannot be
reached by email anymore. I have already posted messages to NOBULL and
AREIBENS because their secret email addresses bounce.

Messages do not expire ever, they need to be deleted once they are
outdated, preferably by the admin who wrote the message, but other
admins can jump in if needed. But the receipients cannot delete the
messages.

So the PAUSE has two new pages in the admin menu: post a message,
delete messages.

Messages are limited to 255 characters to keep the annoyance to the
users low.

For better transparence I have, at the same time, improved on the
"switch user" feature. When you click "Select User/Action" and do
something on behalf of a user, the status widget on the top right
always shows both your name and email and the other user's name and
email. And it also shows the pending messages from the message board
*for the other user*.

Coincidentally, the "Delete messages" page has already prepared links
to these pages, just to keep potentially useful clicking paths short.
If you want to see this in action, the quickest way is to post a
message to yourself.

Enjoy,
-- 
andreas

Re: Request for Namespace XML::SimpleObject - XML and Objects

[Andreas J. Koenig]

> On Mon, 24 Feb 2003 04:07:32 +0100 (CET), Tina Mueller <[EMAIL PROTECTED]> said:

  > I would be glad about a suggestion for a namespace
  > which I could use for my module.

Maybe try the XML mailing list? [EMAIL PROTECTED]

Or ask Grant McLean directly where he sees fit for your code.

-- 
andreas

Re: REQ: rename MPEG::ID3v1Tag to MP3::ID3v1Tag on modules list.

[Andreas J. Koenig]

> On Fri, 21 Feb 2003 11:47:27 +0100, "A. Bergman" <[EMAIL PROTECTED]> said:

 >> I sent in this request years ago, and I was wondering if we could
 >> make this adjustment in the modules list. I have removed the backwards
 >> compatibilty wrapper from the latest release.

  > I am not really sure how to do it, but if someone tells me I will :-)

There's no way to rename a module via PAUSE except for registering a
new module and changing the status of the old one to "can be deleted".

So I just did it to the mysql database directly. As I do not expect
many cases like that in the future, I think there's no need to tell
you how it can be done.

-- 
andreas

Re: Fw: PAUSE Indexer report I/IN/INGY/only-0.10.tar.gz

[Andreas J. Koenig]

>>>>> On Tue, 18 Feb 2003 13:09:04 -0800, Brian Ingerson <[EMAIL PROTECTED]> said:

  > On 18/02/03 21:54 +0100, Andreas J. Koenig wrote:
 >> >>>>> On Tue, 18 Feb 2003 11:59:12 -0800, Brian Ingerson <[EMAIL PROTECTED]> said:
 >> 
 >> > I think that PAUSE should ignore everything under t/.
 >> 
 >> So true! Fixed in rev 188

  > So does this mean I don't need to fix anything? Will only.pm get through
  > the system as is?

Yes, it did already.

  > I see the number '188' in there, so I assume its all good.

Yes, all is good:-)

perl -le '
use only only;  
use if only => only;  
use only only => print("ok, Good night")/10 ;
'
ok, Good night



-- 
andreas

Re: Fw: PAUSE Indexer report I/IN/INGY/only-0.10.tar.gz

[Andreas J. Koenig]

> On Tue, 18 Feb 2003 11:59:12 -0800, Brian Ingerson <[EMAIL PROTECTED]> said:

  > I think that PAUSE should ignore everything under t/.

So true! Fixed in rev 188

Thanks,
-- 
andreas

Re: Scheduled for reindexing

[Andreas J. Koenig]

> On Sun, 16 Feb 2003 10:57:49 -0800, Andrew Zhilenko <[EMAIL PROTECTED]> said:

  > Hello!
  > Perl Authors Upload Server wrote:

  > I have submitted updated version of Authen::Radius module on the
  > February, 7th. And, since it still was not available on the
  > www.cpan.org, I have requested reindexing yesterday. Yes, it is still
  > not available there. Am I doing something wrong?

http://use.perl.org/articles/03/02/16/1416231.shtml?tid=32

-- 
andreas

Re: Where did the latest modules go?

[Andreas J. Koenig]

> On Sun, 16 Feb 2003 12:17:47 +0100, Gunnar Hjalmarsson <[EMAIL PROTECTED]> said:

  > Graham Barr wrote:
 >>> Uploads later than 10 February are no longer available in CPAN,
 >>> which seems to be out of sync with PAUSE.
 >> grep -c "deleting authors" var/log/rsync.log.20030214161500
 >> 203
 >> That does seem a lot to be deleting.
 >> The two distributions I have had reported are Acme-Time-Baby and
 >> Filesys-Ext2
 >> both of which were deleted by the above rsync run.

  > I believe it's a lot more than that. When searching or browsing in
  > CPAN, I don't see *any* modules uploaded later than 10 February.

  > My own module is CGI::ContactForm. It shows version 0.2, 0.3 and 0.4,
  > even if all those should have been deleted via PAUSE. Version 1.0
  > (scheduled for deletion) and 1.01, uploaded 12 respective 13 February,
  > are not visible.

Confirmed. ftp.funet.fi, the CPAN master site, has a RAID problem
since over a week and is running in degraded mode. Apparently they
have decided to not run the mirror process anymore until the problem
is fixed.

I don't know how long it will take to recover the disks in Finland. I
guess we should better wait another few days before we start
redesigning the whole CPAN mirroring tree.


-- 
andreas

Re: PAUSE Indexer report P/PA/PARDUS/zoidberg/Zoidberg-0.2.tar.gz

[Andreas J. Koenig]

Raoul,

the appended report seems to indicate that you are using a multitude
of namespaces. Maybe the report is incorrect, I have not checked all
details, but a cursory look finds a lot of root namespace pollution.

I hope you do not feel torpedized in this sophisticated project when I
ask you to clean that up a little. Thank you!

-- 
andreas

> On Wed, 12 Feb 2003 18:25:29 +0100, PAUSE <[EMAIL PROTECTED]> said:

  > The following report has been written by the PAUSE namespace indexer.
  > Please contact [EMAIL PROTECTED] if there are any open questions.
  >   Id: mldistwatch 172 2003-02-11 22:28:43Z k 

  > CPAN ID: PARDUS
  >   Distribution file: P/PA/PARDUS/zoidberg/Zoidberg-0.2.tar.gz
  > Number of files: 65
  >  *.pm files: 25
  >   Timestamp of file: Wed Feb 12 17:14:18 2003 UTC
  >Time of this run: Wed Feb 12 17:25:29 2003 UTC

  > The following packages have been found in the distro:

  > String
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/Fish/Prompt.pm
  >  status: Not indexed because permissions missing. Visit PAUSE and
  >  click "View Permissions" to find the legitimate
  >  maintainer(s).

  > String::Scrolling
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/Fish/Prompt.pm
  >  status: Not indexed because Zoidberg/Prompt/Prompt.pm in
  >  P/PA/PARDUS/zoidberg/Zoidberg-0.1c-unstable.tar.gz has a
  >  higher version number (0.1)

  > Zoidberg::Fish
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/Fish.pm
  >  status: Not indexed because Zoidberg/Fish/Fish.pm in
  >  P/PA/PARDUS/zoidberg/Zoidberg-0.1c-unstable.tar.gz has a
  >  higher version number (0.1)

  > Zoidberg::PdParse
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/PdParse.pm
  >  status: Not indexed because Zoidberg/PdParse/PdParse.pm in
  >  P/PA/PARDUS/zoidberg/Zoidberg-0.1c-unstable.tar.gz has a
  >  higher version number (0.1)

  > Zoidberg::StringParse
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/StringParse.pm
  >  status: Not indexed because Zoidberg/StringParse/StringParse.pm in
  >  P/PA/PARDUS/zoidberg/Zoidberg-0.1c-unstable.tar.gz has a
  >  higher version number (0.1)

  > Zoidberg::StringParse::Syntax
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/StringParse/Syntax.pm
  >  status: Not indexed because Zoidberg/StringParse/Syntax/Syntax.pm
  >  in P/PA/PARDUS/zoidberg/Zoidberg-0.1c-unstable.tar.gz has a
  >  higher version number (0.1)

  > fortune
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/Fish/MOTZ.pm
  >  status: Not indexed because Zoidberg/MOTZ/MOTZ.pm in
  >  P/PA/PARDUS/zoidberg/Zoidberg-0.1c-unstable.tar.gz has a
  >  higher version number (0.1)

  > Makefile
  > version: 0.1
  > in file: Zoidberg-0.2/m/Makefile.pm
  >  status: indexed

  > Scuddle
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/ZoidParse.pm
  >  status: indexed

  > Scuddle::Native
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/ZoidParse.pm
  >  status: indexed

  > Scuddle::Wide
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/ZoidParse.pm
  >  status: indexed

  > Zoidberg
  > version: 0.2
  > in file: Zoidberg-0.2/lib/Zoidberg.pm
  >  status: indexed

  > Zoidberg::Eval
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/Eval.pm
  >  status: indexed

  > Zoidberg::FileRoutines
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/FileRoutines.pm
  >  status: indexed

  > Zoidberg::Fish::Buffer
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/Fish/Buffer.pm
  >  status: indexed

  > Zoidberg::Fish::Buffer::Insert
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/Fish/Buffer/Insert.pm
  >  status: indexed

  > Zoidberg::Fish::Buffer::Insert::MultiLine
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/Fish/Buffer/Insert/MultiLine.pm
  >  status: indexed

  > Zoidberg::Fish::Buffer::Insert::VimCommand
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/Fish/Buffer/Insert/VimCommand.pm
  >  status: indexed

  > Zoidberg::Fish::Buffer::Meta
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/Fish/Buffer/Meta.pm
  >  status: indexed

  > Zoidberg::Fish::Buffer::Meta::Vim
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/Fish/Buffer/Meta/Vim.pm
  >  status: indexed

  > Zoidberg::Fish::Buffer::Select
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/Fish/Buffer/Select.pm
  >  status: indexed

  > Zoidberg::Fish::Commands
  > version: undef
  > in file: Zoidberg-0.2/lib/Zoidberg/Fish/Commands.pm
  >  status: indexed

  > Zoidb

Re: PAUSE Indexer report E/EC/ECALDER/cpan2rpm-2.011.tar.gz

[Andreas J. Koenig]

> On Tue, 11 Feb 2003 07:56:38 -0800, "Erick Calder" <[EMAIL PROTECTED]> said:

  > what does this mean and how do I fix it?

The report is misleading, sorry for that. I'll try to change the
script that creates it.

What it means is that cpan2rpm-2.011.tar.gz does not contain any
modules and as such not any package statements. As it is a tarball
around a script, this is exactly what you intended and there's nothing
to fix on your side.

  > -Original Message-
  > From: PAUSE [mailto:[EMAIL PROTECTED]]
  > Sent: Tuesday, February 11, 2003 2:24 AM
  > To: [EMAIL PROTECTED]; [EMAIL PROTECTED]
  > Subject: PAUSE Indexer report E/EC/ECALDER/cpan2rpm-2.011.tar.gz


  > The following report has been written by the PAUSE indexer.

I'll try s/indexer/namespace indexer/

  > Please contact [EMAIL PROTECTED] if there are any open questions.

  >   Id: mldistwatch 170 2003-02-11 07:08:07Z k 

  > CPAN ID: ECALDER
  >   Distribution file: E/EC/ECALDER/cpan2rpm-2.011.tar.gz
  > Number of files: 10

I'll add a line counting *.pm files.

  >   Timestamp of file: Tue Feb 11 10:05:56 2003 UTC
  >Time of this run: Tue Feb 11 10:23:43 2003 UTC

  > No packages could be identified in the distro. Nothing done.

I'll try instead if there is at least one *.pm file:

  No package statements could be found in the distro (maybe a script
  or documentation distribution?)

and will not send a report at all if there is no *.pm file.

Other suggestions welcome.

-- 
andreas

Re: ranting about retrograde again

[Andreas J. Koenig]

> On 16 Jan 2003 13:30:51 -0800, [EMAIL PROTECTED] (Randal L. Schwartz) said:

  > 

Your prayers have been heard:-)

PAUSE's indexer since yesterday sends a report about each distro's
indexing to the author.

It sends version parsing information about all found packages and
specifically it complains about

-- not a single root directory when unpacking,
-- distribution contains a blib directory,
-- falling version numbers.

I hope your retrograde rant won't be that often necessary anymore.

If there are other things we should point out in this letter to the
authors, please let me know.

-- 
andreas

Re: naming of map projection module

[Andreas J. Koenig]

> On Mon, 10 Feb 2003 18:53:16 -0900, Dan Stahlke <[EMAIL PROTECTED]> said:

  > Greetings, Modules Group:
  > I am about to upload a module that is a wrapper around the GCTPc
  > library.  What this module does is convert coordinates from one map
  > projection to another.  It is currently named Cartography::Projection,
  > but I am thinking perhaps it should be renamed to Cartography::GCTP so
  > that it would be named after the underlying C library.  The advantage
  > is that the name would not clash with future projection libraries that
  > people may write, the disadvantage is that not everyone knows what
  > GCTP is.  Do you have any thoughts on this?

I prefer Cartography::GCTP because we may expect that cartographers
know about the library; and wannabee cartographers should read the
READMEs of all modules in the Cartography namespace.

-- 
andreas

Re: PAUSE Indexer report A/AB/ABIGAIL/Regexp-Common-2.109.tar.gz

[Andreas J. Koenig]

> On Tue, 11 Feb 2003 01:29:25 +0100, Abigail <[EMAIL PROTECTED]> said:

 >> Regexp::Common::Entry
 >> version: 2.109
 >> in file: Regexp-Common-2.109/lib/Regexp/Common.pm
 >> status: Not indexed because permissions missing. Visit PAUSE and
 >> click "View Permissions" to find the legitimate
 >> maintainer(s).


  > What does this mean? I've updated my versions of Regexp::Common,
  > and all used a package named Regexp::Common::Entry buried somewhere
  > in the code. Except for the version number, nothing has changed in
  > in the file lib/Regexp/Common.pm for many updates of the package.

  > Why now this report? What did I do wrong, and how should I fix it?

I just fixed it for you. What happened was that Damian still was
registered as the primary maintainer of Regexp::Common::Entry. I
deleted that record and ran the indexer again against 2.109. The
indexer then registered you as the maintainer.

The reason why you never received such reports before is that PAUSE
did not have them before yesterday. They are now there to shake
exactly such little bugs out of the index.

-- 
andreas

Re: CPAN daily update list

[Andreas J. Koenig]

> On Sun, 9 Feb 2003 20:38:30 +1100, Ron Savage <[EMAIL PROTECTED]> said:

  > Folks
  > My module uploads don't seem to be ever listed on the daily list.

I only checked DBIx-Hash2Table-1.02.tgz and found that it contains a
blib/ directory. This is interpreted by the PAUSE indexer as a binary
distribution that is not to be indexed.

-- 
andreas

Re: CPAN submission guidelines

[Andreas J. Koenig]

Thank you, Arthur, for voicing your frustration. We do not have a
bullet proof concept how to avoid frustration, but what happened to
you seems avoidable.

> On Fri, 7 Feb 2003 08:27:07 -0900 (AKST), Arthur Corliss 
><[EMAIL PROTECTED]> said:

  > Greetings:
  > Let me start by saying I know you guys have day jobs as well, but I have to
  > wonder if I'm following the right procedure on getting modules registered on
  > CPAN.  My last couple of attempts appear to be failing miserably.

  > First off, my understanding about namespace registration was to submit the
  > standard form and wait for feedback.  I did that with Parse::PlainConfig.  As
  > per 04pause.html:

  >   The module list maintainers themselves are mostly lurkers. You need not wait
  >   for a response. Generally a lack of response can be taken as acceptance of
  >   the module name being proposed.

Well, at least there it really says: You need *not* wait. You get
maintainer status by simply using a namespace the first time.

  > I want to point out that Parse::PlainConfig as a name was actually proposed by
  > one of list maintainers.  I accepted the recommendation, and based on the
  > 04pause.html doc, thought that module was accepted for inclusion.

I'm reading the old thread again and I think the problem started with
the namespace Parse::PerlConfig. Tim said:

  Perhaps Parse::PlainConfig. Seems more natural to me that 'RCfile',
  especially given the existance of Parse::PerlConfig.

Parse::PerlConfig is not registered and imho is a bad precedent.
Parse:: modules should deal with parse algorithms: Lex, RecDescent,
Tokens, YALALR, Yapp are dealing with the task of parsing in a generic
sense. I agree with brian that modules for reading config files should
live in the Config or ConfigReader namespace.

  > A month ago, I find out from brian d foy that my module was
  > *never* registered officially on CPAN, and yet I've already
  > polluted that namespace. And since he's opposed to the name, I
  > know I have no chance of getting it registered under that name.

Oh well, I think, I'm ready to break down. Not by the namespace choice
but by the fact that Tim's statement looked pretty approving and quite
a lot of time has passed since. It's not acceptable for the users of
your module to now reject a namespace proposal such a long time after
it seemed to be approved.

If nobody objects within the next three days, I'll approve
Parse::PlainConfig.

  > Fast forward to the last few weeks.  I have a new module that I'd like to
  > register (Class::Aggregate).  I'll warrant that I didn't fully research out my
  > options before proposing it (and I thank all of you that pointed me in the
  > right direction), but since then I've found that none of the existing modules
  > a) fit my feature criterion, and b) can easily be combined with other modules
  > to do so without considerable overhead, either.

  > After some discussion, at least one of you seemed open to having it uploaded
  > to CPAN, but wanted a name change.  I can't think of anything better, so I
  > took a recommendation from you guys, once again.

  > What I'm finding is that even if I accept your recommendations, that namespace
  > is *still* not officially registered on CPAN, nor can I update module
  > metadata via the pause interface.

If you mean Class::EHierarchy, I've registered that now.

  > I'm loathe to upload anything you guys don't approve of out of respect for the
  > archive.  I don't want to hijack any namespace that the community doesn't feel
  > is appropriate.

  > So, what am I doing wrong?  Can I get modules registered, or am I just
  > supposed to go away?  I apologise if any of you feel that I'm being a pain in
  > the ass.  Either tell me to go away (which I will comply with), or give me a
  > procedure that will work.

CPAN is extremely tolerant in what it accepts. It works without human
intervention. First-come-first-served is the default. It's the extra
blessing of being registered in the module list that needs human
intervention and this is a painful bottleneck. We do not know how this
can be improved.

You could perhaps adjust your expectation: ask us for advice and feel
free *not* to accept it. The number of CPAN developers who either
ignore us completely or do not follow our suggestions or did not get
any good suggestions from us seem to be the majority, and yet CPAN
works mostly.


-- 
andreas

Re: Registering myself

[Andreas J. Koenig]

> On Thu, 06 Feb 2003 11:50:54 -0800, Craig Lawson <[EMAIL PROTECTED]> 
>said:

  > Hi Folks,
  >   Just a reminder: I never received an account.

 >> Hi PAUSE Maintainers,
 >> I'd like to register myself as a Perl author.
 >> Name:Craig Lawson
 >> E-mail:  [EMAIL PROTECTED]
 >> Home page:   http://pages.sbcglobal.net/craig.lawson
 >> CPAN ID: CLAW

You have been registered on 2002-10-22 with these data:

CLAW
[EMAIL PROTECTED]
http://pages.sbcglobal.net/craig.lawson
Craig Lawson

Pause's sendmail log says:

Oct 22 23:48:03 pause sendmail[4418]: g9MLm1704416: [EMAIL PROTECTED]
t, ctladdr=nobody (99/99), delay=00:00:02, xdelay=00:00:02, mailer=esmtp, pri=31
226, relay=vmc-ext.prodigy.net. [207.115.63.88], dsn=2.0.0, stat=Sent (g9MLm2Vv2
312098 Message accepted for delivery)

Please try the "Forgot Password" button on PAUSE and see if you get a
password sent. I hope that resolves the issue. If not, please let us
know.

-- 
andreas

Re: Archive::Tar ownership

[Andreas J. Koenig]

> On Thu, 06 Feb 2003 17:13:01 +0100, Jos Boumans <[EMAIL PROTECTED]> said:

  > Greetings,
  > the Archive::Tar module has been unmaintained for the last 3 years
  > now, and seeing that it's one of the modules CPANPLUS (of which I am
  > the author) relies upon in a non-Unix environment, it's very
  > unhelpfull that it's broken on exactly those platforms =/

  > I've sent the author patches to fix this but he's been unresponsive.
  > I've taken the liberty to upload a 0.23 version to the CPAN which
  > falls back to an older version of A::T when not on unix platforms to
  > circumvent the problem, but I'd like to get a /proper/ fix in.

  > So, hereby I so formally request to be given ownership of the
  > Archive::Tar namespace, in the hopes to fix things so they May Work
  > Again.

Thank you for stepping forward. I'm CC-ing this to Stephen, the
current maintainer and Schwern because I seem to recall that Schwern
sent Stephen patches. I also sent patches to Stephen which I cannot
find anymore. Maybe Stephen, if you still have them, could you send
the patches to Jos to integrate them?

I ask y'all if there are any objections to passing maintainership to
Jos. If there are no objections till Wednesday, I'll take the
necessary steps on PAUSE to allow Jos official uploading for
Archive::Tar.

Thanks again,
-- 
andreas

Re: Need to change my info, upload stuff...

[Andreas J. Koenig]

> On Thu, 6 Feb 2003 00:19:55 -0800, Sam Corl <[EMAIL PROTECTED]> said:

  > Hi,
  > I just realized that my Author account at CPAN is outdated.  Please
  > send instructions on how to update my contact information and submit
  > new modules.

  > thanks in advance for any info. you can provide...

If the stored email address is outdated you can only tell us which
email adress it should be from now on. Otherwise just visit

http://pause.perl.org

and click on "Forgot password".


-- 
andreas

Re: Module submission Class:Aggregate

[Andreas J. Koenig]

> On Mon, 3 Feb 2003 11:11:03 -0900 (AKST), Arthur Corliss 
><[EMAIL PROTECTED]> said:

  > On Tue, 28 Jan 2003, Henrik Tougaard wrote:
 >> When I saw the modulename I immediately thought of a 'has-a' relationship.
 >> In UML that is called Agrregation.
 >> The name ought to convey sa better idea of what the module does -
 >> Class::Aggregate is IMHO just too generic. I can't think of a better name
 >> though.

  > I know, that's my quandry as well.  Let's face it:  there's no reasonably
  > short name that's going to be desriptive enough.  So, if not that, how about:

  >   Class::Lite  (since it's a relatively light-weight implementation)
  >   Class::Orderly  (referencing the point of the container relationship
  > for destruction purposes)

  > Hell, call it Class::Foo, for all I care, I'd just like to get an officially
  > sanctioned module that does what I want and can be reused across all the
  > Curses::* modules I'm working on.  I'm getting tired of duplicating code. . .

  > Any ideas?

Before you call it Foo or Bar or Lite, try something that resembles
your intent even if not being really descriptive. Orderly looks fine
to me as do the following:

Class::Event
Class::Eventable
Class::Hierarchy
Class::EHierarchy
Class::ECHierarchy



-- 
andreas

Re: spam filters

[Andreas J. Koenig]

> On Mon, 3 Feb 2003 04:54:36 -0800 (PST), Ask Bjoern Hansen <[EMAIL PROTECTED]> said:

  > We should get less spam to [EMAIL PROTECTED] and [EMAIL PROTECTED] now
  > (and to most of the other lists too).

Thank You! That's highly appreciated.

-- 
andreas

Re: Namespace-Permission Problem Fw: New module Class::Prevayler

[Andreas J. Koenig]

> On Fri, 24 Jan 2003 11:04:38 -0500, Kurt Starsinic <[EMAIL PROTECTED]> said:

 >> The module Class::Prevayler isn't available on CPAN.
 >> 
 >> Either the module has not yet been uploaded to CPAN, or it is
 >> temporary unavailable. Please contact the author to find out
 >> more about the status. Try 'i Class::Prevayler'.
 >> 
 >> I'm the author, but i don't know what the status is ? :-)

  > Something's funny about the tarball.  It's in CPAN, and you can
  > download it by filename, but PAUSE isn't recognizing it as containing
  > Class::Prevayler.  Andreas?

Fixed. -- A bug I introduced in rev. 117 partially obstructed the
indexing. In rev 129 I have now fixed the bug and the index is
repaired.

Please let me know if anything still looks odd.

Thanks for the report!
-- 
andreas

Re: Module submission User::Identity

[Andreas J. Koenig]

Now it looks like a bikeshed problem to me.

http://www.freebsd.org/doc/en_US.ISO8859-1/books/faq/misc.html#BIKESHED-PAINTING

Sorry about that.

> On Tue, 21 Jan 2003 11:09:45 +0100, Mark Overmeer <[EMAIL PROTECTED]> said:

  > For my feeling, the focus of all responses till now is that the main issue
  > about user information is that it is stored in an address book.  For me, the
  > main issue is that it is easily accessible and smart.

If an Addressbook::Entry object is not easily accessible and smart it
is clearly not an option. Otherwise it might be good for reuse.

  > A User::Indentity combines information from many sources for a temporal
  > existence.  Calling the module anything which implies that it is persistent
  > information is wrong.  Calling it anything which restricts itself to a
  > single source of data is a mistake as well.

The name Addressbook::Entry does not necessarily imply persistency or
single data source. It's just a metapher everybody understands.

-- 
andreas

Re: Module submission User::Identity

[Andreas J. Koenig]

> On Tue, 21 Jan 2003 08:50:35 +0100, Mark Overmeer <[EMAIL PROTECTED]> said:

  > Please suggest something which is more suitable.

Have you had a look at AddressBook::Entry? I haven't, but I believe a
top level namespace like AddressBook or Rolodex would characterize the
role of the module better than a namespace in the Person/User/Identity
realm.

-- 
andreas

Re: CGI::WML Perms Update

[Andreas J. Koenig]

>>>>> On Fri, 17 Jan 2003 19:54:52 -0500, Andy Murren <[EMAIL PROTECTED]> said:

  > Andreas,
  > On Tue, Jan 14, 2003 at 03:31:28PM +0100, Andreas J. Koenig wrote:
  > : >>>>> On Mon, 13 Jan 2003 15:43:01 -0500, Andy Murren <[EMAIL PROTECTED]> said:
  > : 
  > :   > I am the co-author on CGI::WML,  Angus Wood is the primary author and
  > :   > the only one with permission to update the module.  I have not been
  > :   > able to contact him in almost 2 years.  How can I get permission to
  > :   > take over maintaining the Module?
  > : 
  > : Angus? Ping?
  > : 
  > : -- 
  > : andreas

  > How long do you think we should wait?  Could you just add me instead
  > of replacing Angus?

I just did so.


-- 
andreas

Re: [perl #20397] When will Quantum::Superpositions be turned over to me? (fwd)

[Andreas J. Koenig]

> On Mon, 20 Jan 2003 10:57:19 -0600, _brian_d_foy <[EMAIL PROTECTED]> said:

 >> Steven was victim of the same bug I fixed on Sunday. Since then this
 >> problem can be solved by reindexing. 

  > is that something users can do themselves with the "Reindex" item on
  > the PAUSE menu, or is it a different reindexing?

It's the one on the PAUSE menu, there is no other kind of reindexing.

-- 
andreas

Re: PAUSE ID request (FENGHONGXIA; fenghongxia)

[Andreas J. Koenig]

> On Mon, 20 Jan 2003 03:08:23 +0100, "Perl Authors Upload Server" 
><[EMAIL PROTECTED]> said:

  > Request to register new user
  > fullname: fenghongxia
  >   userid: FENGHONGXIA
  > mail: CENSORED
  > homepage: 
  >  why:

Somebody seems to prove that there is a bug in the
ID-conformance-enforcement step. I'll look into it.

-- 
andreas

Re: [perl #20397] When will Quantum::Superpositions be turned over to me? (fwd)

[Andreas J. Koenig]

> On Sun, 19 Jan 2003 12:33:59 -0600, _brian_d_foy <[EMAIL PROTECTED]> said:

  > In article <20354.1042978791@[192.168.200.4]>, Steven Lembark
  > <[EMAIL PROTECTED]> wrote:

 >> Somebody, what does it take to have the current verison of
 >> Quantum::Superpositions made available (i.e., not the one
 >> in Damian's directory but my own)? The release is at 1.05
 >> but noone can get to it!

 cpan> i Quantum::Superpositions
 >>> CPAN: Storable loaded ok
 >>> Going to read /home/lembark/.cpan/Metadata
 >>> Database was generated on Sat, 18 Jan 2003 14:48:40 GMT
 >>> Strange distribution name [Quantum::Superpositions]Module id =
 >>> Quantum::Superpositions DESCRIPTION  QM-like superpositions in Perl
 >>> CPAN_USERID  DCONWAY (Damian Conway <[EMAIL PROTECTED]>)
 >>> CPAN_VERSION undef

  > it looks like CPAN_VERSION is the problem.  CPAN.pm does not
  > know which version is which.  the package defines a $VERSION,
  > so i do not see why CPAN.pm wouldn't see it.

  > the module is registered to you in the Modules List, but PAUSE
  > still thinks it belongs to Damian (according to "View Permissions").

  > Andreas, is this an indexing problem?

Steven was victim of the same bug I fixed on Sunday. Since then this
problem can be solved by reindexing. New uploads won't have it
anymore. I do not know how many uploads before Sunday are affected.
They can be identified in that "View Permissions" shows a
co-maintainer but no maintainer.

Offhand I do not see a simple SQL statement or perl program that finds
all of them. If it comes up more often, I'll think about it some more.

Sorry, Steven, for the bug. You're now officially blessed as the
maintainer. It just needs a day to propagate through the CPAN.


-- 
andreas

Re: WWW::Search::Scraper::NorthernLight went retrograde

[Andreas J. Koenig]

> On Fri, 17 Jan 2003 22:52:38 -0800, Glenn Wood <[EMAIL PROTECTED]> 
>said:

  > I'm working on it . . . bundling these modules makes more sense
  > anyway, so perhaps that's the way I'll go (or do I have to own a
  > namespace to do that? hmm . . .)

On CPAN, as it was accidentally designed, ownership springs into
existence by using an namespace for the first time. So, in the normal
case you need not worry about ownership unless you use a namespace
somebody else has already taken. If a namespace is already in use can
easily be checked by

- asking the CPAN.pm shell
- grepping through $CPAN/modules/02packages.details.txt.gz
- Clicking on "View permissions" on PAUSE


-- 
andreas